Production Planning and Control in an Ambulatory Care Service Provider in
Sweden
CARL THELIN PETER WALLANDER
Master of Science Thesis
Stockholm, Sweden 2015
ii
Produktionsplanering i en ambulant vårdgivare i Sverige
CARL THELIN PETER WALLANDER
Examensarbete
Stockholm, Sverige 2015
i
Produktionsplanering i en ambulant vårdgivare i Sverige
av
Carl Thelin Peter Wallander
Examensarbete INDEK 2015:83
KTH Industriell teknik och management
Industriell ekonomi och organisation
SE-100 44 STOCKHOLM
ii
Production Planning and Control in an Ambulatory Care Service Provider in
Sweden
Carl Thelin Peter Wallander
Master of Science Thesis INDEK 2015:83
KTH Industrial Engineering and Management
Industrial Management
SE-100 44 STOCKHOLM
iii
s
Examensarbete INDEK 2015:83
Produktionsplanering i en ambulant vårdgivare i Sverige
Carl Thelin
Peter Wallander
Godkänt
2015-06-19
Examinator
Mats Engwall
Handledare
Anna Jerbrant
Uppdragsgivare
Danderyds sjukhus AB
Kontaktperson
Hans Lindgren
Sammanfattning
Sjukvården i Sverige står inför flertalet stora utmaningar i dag. Med en växande och åldrande
befolkning ökar vårdbehovet i både komplexitet och volym. Samtidigt är resurserna begränsade
vilket leder till att sjukvården måste bli mer effektiv i sitt vårderbjudande. Sjukvården har därför
börjat titta på management-teorier från andra branscher, främst bilindustrin. Ambulanta
vårdgivare, enheter som utför vårdtjänster utan egna inneliggande patienter, är centrala i
produktivitets och effektiviseringsförbättringar i de system de agerar i. Denna studie har undersökt
hur forskningens idéer kring produktionsplanering och kontroll kan användas av en ambulant
vårdgivare.
Detta examensarbete är baserat på en fallstudie genomförd på röntgenavdelningen på Danderyds
sjukhus i Stockholms län. Fallstudien har utgjorts av framförallt kvalitativ datainsamling genom
semistrukturerade intervjuer understödda av kvantitativ produktionsdata och en
benchmarkingstudie på röntgenmottagningen på Universitetssjukhuset i Linköping. Det empiriska
bidraget i fallstudien har inkluderat en kartläggning av arbetsprocessen i röntgenmottagningen på
Danderyds sjukhus. Processen kunde delas upp i tre faser: 1. Förberedelser, 2. Undersökning, och
3. Diagnosticering och dokumentering.
Utifrån den kartlagda processen och en utförlig litteraturstudie om produktionsplanering och
kontroll kunde tre huvudsakliga slutsatser dras: (1) implementering av produktionsplanering och
kontroll måste utgå från en grundlig förståelse för processerna hos vårdgivaren och syfta till att
verka för både kliniska och operationella mål, (2) produktionsplanerings- och kontrollaktiviteter
måste anpassas till vårdgivarens omgivning, och (3) kontinuerlig återkoppling från kvalitets och
produktivitetsmål är oumbärliga för framgångsrik utnyttjning av produktionsplanerings och
kontroll i en ambulant vårdgivare.
Detta examensarbete kommer hjälpa ambulanta vårdgivare att möte de utmaningar och
produktivitetskrav dessa står inför i Sverige genom att utnyttja potentialen med
produktionsplanering och kontroll.
Nyckelord: Produktionsplanering och kontroll, operations management, sjukvård,
ambulant vårdgivare, röntgen
iv
Master of Science Thesis INDEK 2015:83
Production Planning and Control in an Ambulatory Care Service Provider in Sweden
Carl Thelin
Peter Wallander
Approved
2015-06-19
Examiner
Mats Engwall
Supervisor
Anna Jerbrant
Commissioner
Danderyds sjukhus AB
Contact person
Hans Lindgren
Abstract The healthcare sector in Sweden faces several challenges today: a growing and ageing population
increases demand in terms of both volume and complexity whereas monetary resources available
do not follow suit, forcing care givers to become more efficient in their operation. The healthcare
sector has for this reason in recent years become more and more inclined to implement
management theories developed in other industries, primarily the automotive industry.
Ambulatory care service units, who provide care services on an outpatient basis, are vital in
productivity improvements in the system they are acting in but have been somewhat neglected by
both policy makers and academia. This study has therefore investigated how production planning
and control theories, principles and methods can be utilised in an ambulatory care service unit in
a major emergency hospital in Sweden.
This thesis is based on a case study conducted at the radiology department at Danderyds sjukhus,
an emergency hospital located in Stockholm County Council. The case study mainly consisted of
qualitative data gathering using semi-structured interviews with aid from quantitative data on
department performance. The empirical contributions of the case study included a process
mapping of a generic process flow prevalent in all subunits of the department in the radiology
department. The main process could be divided into three main phases: 1. Preparation, 2.
Examination, and 3. Diagnosing and documentation.
Using the mapped out process and a thorough literature review on production planning and control
three main conclusions could be drawn: (1) an implementation of production planning and control
should begin with a deep understanding of process flows in the unit and aim to promote both
operational and medical objectives, (2) production planning and control activities should be
adapted to the environment the unit is acting in, and (3) continuous feedback from performance
measurements is vital to successful production planning and control initiatives.
This study will help ambulatory care service units to meet the challenge of increased demand they
currently face in Sweden by utilising the potential in production planning and control.
Key-words: Production planning and control, operations management, healthcare,
ambulatory care services, radiology
v
Table of Contents
Table of Figures ............................................................................................................................. vi
Tables ............................................................................................................................................. vi
Glossary and Acronyms ................................................................................................................ vii
Acknowledgements ...................................................................................................................... viii
1 INTRODUCTION ........................................................................................................................ 1
1.1 PROBLEM FORMULATION .................................................................................................. 2
1.2 PURPOSE ............................................................................................................................ 2
1.3 RESEARCH QUESTIONS ...................................................................................................... 4
1.4 DELIMITATIONS ................................................................................................................. 4
1.5 THESIS OUTLINE ................................................................................................................ 4
2 METHOD .................................................................................................................................. 6
2.1 RESEARCH DESIGN ............................................................................................................ 6
2.2 ANALYSIS OF EMPIRICAL DATA......................................................................................... 9
3 LITERATURE AND THEORY ................................................................................................... 12
3.1 PRODUCTION PLANNING AND CONTROL .......................................................................... 13
3.2 PRODUCTION PLANNING AND CONTROL IN HEALTHCARE ............................................... 20
4 EMPIRICAL SETTING ............................................................................................................. 26
4.1 DANDERYDS SJUKHUS ..................................................................................................... 26
4.2 RADIOLOGY DEPARTMENT AT DANDERYDS SJUKHUS ..................................................... 27
4.3 RADIOLOGY DEPARTMENT AT UNIVERSITETSSJUKHUSET I LINKÖPING ........................... 34
5 RESULTS ................................................................................................................................ 35
5.1 THE GENERAL PROCESS AT THE RADIOLOGY DEPARTMENT ............................................ 35
5.2 CONTINGENCIES OF DIFFERENT MODALITIES .................................................................. 40
5.3 BENCHMARK STUDY ........................................................................................................ 49
6 ANALYSIS .............................................................................................................................. 51
6.1 PRODUCTION PLANNING AND CONTROL IN AN AMBULATORY CARE SERVICE PROVIDER 51
6.2 PROCESS FLOW IN RADIOLOGY DEPARTMENT ................................................................. 57
6.3 COLLABORATION AND COORDINATION ........................................................................... 62
7 CONCLUSIONS ....................................................................................................................... 66
7.1 ANSWERS TO RESEARCH QUESTIONS ............................................................................... 66
7.2 CONCLUDING REMARKS .................................................................................................. 70
8 BIBLIOGRAPHY ...................................................................................................................... 73
vi
TABLE OF FIGURES
Figure 1: Research Design .............................................................................................................. 6
Figure 2: Number of examinations conducted in the radiology department at Danderyds sjukhus
....................................................................................................................................................... 27
Figure 3: Examinations conducted at different times (percentages, 2014) ................................... 27
Figure 4: Emergency and scheduled examinations (percentages, 2014) ...................................... 28
Figure 5: Examinations conducted over the year for different modalities (percentages, 2014) ... 29
Figure 6: Number of emergency examinations with preliminary response within 2 hours after
incoming referral (2014) ............................................................................................................... 32
Figure 7: Figure 7: Fraction of emergency examinations with preliminary response within 2 hours
after incoming referral ................................................................................................................... 33
Figure 8: Fraction of examinations conducted within 2 hours after arrival .................................. 33
Figure 9: The radiology process fitted into the journey of the patient .......................................... 35
Figure 10: Steps of the Preparation phase ..................................................................................... 36
Figure 11: Steps of the Examination phase ................................................................................... 37
Figure 12: Steps of the Diagnosing and Documentation phase .................................................... 39
Figure 13: Contingencies in the MRI modality ............................................................................. 40
Figure 14: Contingencies in the CR modality ............................................................................... 42
Figure 15: Contingencies in the CT modality ............................................................................... 43
Figure 16: Contingencies in the US modality ............................................................................... 45
Figure 17: Contingencies in the AI modality ................................................................................ 46
Figure 18: Contingencies in the XA modality .............................................................................. 48
TABLES
Table 1: Table of Roles Interviewed in Danderyds sjukhus and at “Universitetssjukhuset i
Linköping” ...................................................................................................................................... 8
vii
GLOSSARY AND ACRONYMS
AI – Angiography
Clinical activities – Activities relating to the bedside treatment of a patient
CR – Computed Radiology
CT – Computed Tomography
DS – Danderyds sjukhus
ERP – Enterprise Resource Planning
Inpatient - A patient who stays in a hospital for one or several nights while given care
IT – Information Technology
JIT – Just-In-Time
MRI – Magnetic Resonance Imaging
MRP – Materials Requirement Planning
MRPII – Manufacturing Resource Planning
Outpatient – A patient who visit a hospital for care without staying
PPC – Production planning and control
TPS – Toyota Production System
US – Ultrasound
XA – Fluoroscopy
viii
ACKNOWLEDGEMENTS
Without certain people it would not have been possible to conduct this thesis. We would like to,
from the bottom of our hearts, thank you for everything.
First and foremost we want to thank our two supervisors; from the academia it is Associate
Professor Anna Jerbrant at the Industrial Economics and Management department at KTH. You
have always been helpful when guiding us through this voyage. From Danderyds sjukhus it is
Hans Lindgren, thank you for all the help you have given us, for always making time to meet
with us and making it possible to get in touch with so many interesting persons to interview.
We would also like to express our appreciation to the personnel at the radiology department at
Danderyds sjukhus. Thank all of you who made time to see us, Viktoria Hägerfelth for all the
help and Johanna Wahlberg for showing great interest in our work and generously contributing
with valuable and astute insights.
Also, thank you Mathias Axelsson at the radiology department at “Universitetssjukhuset i
Linköping”, for taking time to see us. It was a very interesting and fruitful discussion.
Finally, we want to show the uttermost appreciation to our friends and families for everything they
have given to us in our lives, without you we would not be where we are today. We would like to
especially thank Carl’s brother Eric for helping us tremendously with this thesis. Peter’s sister
Karin has also contributed with valuable insights and encouragement for which we are very
grateful. We would also like to thank everyone who through the years have visited CL, you know
who you are and without you KTH would not have been the best time of our lives. Also a thank
you to everyone in Tidningsrummet for all the time spent together, the last couple of years at
KTH became a wonderful time thanks to you. The crew may have changed over time but the room
is bigger than us.
1
1 INTRODUCTION This chapter will give a background to the study and introduce the relevant theoretical area of
production planning and control. It will also present the problem formulation with research
questions and delimitations of the study.
An ever present challenge in healthcare systems is to keep up with a growing demand using scarce
resources, both in terms of funding but also skilled personnel. The expectations in terms of quality
and volume on the systems have been intensified with the potential for better, more complex and
costly treatments. Another pressure on the resource utilisation in healthcare is increased demand
from a growing elderly population all over the Western world, a population that due to high
morbidity is in greater need of care (Molin and Eckerström, 2008; Waring and Bishop, 2010).
Therefore, productivity and efficiency improvements have gained traction in the medical
community already since the 1970s and the implementation of management philosophies in
production industries have become a major focus (Beliën and Demeulemeester, 2008; Brailsford
and Vissers, 2011; Radnor et al., 2012).
An ambulatory care service unit provides healthcare services for patients on an outpatient basis. A
representative example of an ambulatory care service provider is the radiology department in a
hospital, which provides diagnostic services by producing images on internal parts of bodies on
patients referred from other departments. Other examples are primary care services and endoscopy
departments (Hulshof et al., 2012).
Radiological departments face similar challenges as healthcare systems in general, but will also be
further integrated into other hospital care. In addition to those challenges radiological departments
will also have to adapt to the continuously revolutionised medical field of radiology. The
radiological field is nowadays for example interfacing with new and important fields such as
information technology (IT) and molecular biology, as well as a further integration into other
hospital care. With these changes the field will further increase its importance and interest in the
medical community (Chan, 2002; Gill et al., 2005). There are also new radiological technologies
and treatments techniques emerging and with them new diagnostic and treatment possibilities (Gill
et al., 2005). These trends have, in turn, led to the development and application of managerial
techniques to better utilise available resources, eliminate waste and optimise the value of their
services, which has become a major focus for radiological management globally (Khan, 2013;
Kruskal et al., 2012).
Production planning and control can be defined as the planning of processes for manufacturing
and production in a company in order to meet customer demand (Gelders and Van Wassenhove,
1981; Slack et al., 2010). In general, production planning and control has gone from basic shop-
floor planning and labour division in a narrow perspective to a more holistic approach which
requires a deeper understanding of the whole process including the steps that comes before and
after the internal process for an operation (Nguyen et al., 2014; Olhager, 2013).
The process-engineering concept Lean was developed in the automotive industry aiming to reduce
waste in a production system. Organising and designing processes is central in production planning
and control (Slack et al., 2010), which has made Lean a central concept in the theoretical discipline.
Indeed, the interest and application of the concept Lean in healthcare has been widespread globally
in the last decade which has resulted in a fundamentally changed governing dynamics of the
systems (Grufman and Sundblad, 2005; Kruskal et al., 2012; Waring and Bishop, 2010). The Lean
approach can also be said to have a special relevance for radiology in particular because the
practice has been found to depend on a smooth flow of patients (Kruskal et al., 2012).
2
The Swedish healthcare system is primarily financed by the government, constituting
approximately one tenth of government spending, and is invariably of high interest in the society
(Andersson, 2014). Healthcare issues are handled by different county councils. The system is
comparatively, in relation to other Western countries, decentralised with 20 regions governed by
locally elected County Councils responsible for the healthcare in each region (Molin and
Eckerström, 2008).
Danderyds sjukhus, where a case study has been conducted under the scope of this thesis, is a
hospital owned by the Stockholm County Council and funded publicly. The political ownership of
emergency hospitals in Sweden results in a governance that is affected by political trends, which
is apparent at Danderyds sjukhus as well. The increased external demands from the Stockholm
County Council and the government are therefore expected to result in increased strain on resource
utilisation at Danderyds sjukhus.
The population in Stockholm County is growing steadily, and with it the demand for healthcare.
A recent study ordered by the Stockholm County Council suggests that the demand for healthcare
will continue to increase in the coming year. The situation puts a strain on the resources available
for the healthcare system. The trend at Danderyds sjukhus is therefore expected to be similar
(McKinsey & Co, 2013).
The radiology department at Danderyds sjukhus has experienced increased pressure, in terms of
demand volume and complexity, in recent years. The budgetary constraints are raised each year as
the department exceeds the expectations set up in the budget. The department has furthermore
been struggling to cope with demand for both emergency and scheduled examinations within the
desired time frames in the recent year. Employees have also expressed that their working
environment has deteriorated due to stress. Stress levels and pressure on employees are however
hard to measure, given the subjective nature and individual differences. The situation has been
acknowledged by the department head of the radiology department at Danderyds sjukhus.
The radiology department wishes to improve in this situation, thus improving operational
efficiency by utilising production planning and control of production in accordance with demand.
The notion to look into production planning techniques is founded in the implementation of new
production planning systems in the radiology department at “Universitetssjukhuset i Linköping”,
in the beginning of 2012. With the aid of the department head a case study has been conducted at
Danderyds sjukhus to facilitate a thorough investigation of the current situation.
1.1 PROBLEM FORMULATION
Production planning and control in the radiology department at Danderyds sjukhus has proven
challenging due to several aspects. This is partly due to the lack of clearly formulated productivity
and quality goals to be used in a continuous improvement process of the department. There are in
fact no quality measures specifically aimed to radiology departments on a country level in Sweden.
At the same time demand for new examinations has emerged and new referring departments within
the hospital have been established. This combination of external factors contributes to intensified
pressure on the radiology department at Danderyds sjukhus to improve productivity and efficiency.
Improvement work must however not affect quality of care given nor the situation for employees.
Solutions to existing problems must therefore be based on mappings of actual problems from a
production planning and control perspective.
1.2 PURPOSE
3
This thesis is highly relevant as a part of the continuous and radical changes that the healthcare
system in Sweden is going through. New perspectives on accepted problem formulations are
required to facilitate fruitful evolution both within clinical quality measures and operational
governance.
The empirical purpose of this study was to produce a clear mapping of the current situation in the
radiology department at Danderyds sjukhus and to evaluate the department from a production
planning and control perspective. The mapping and evaluation will provide a foundation for further
improvement work in the department regarding production planning and control of the operation
to be able to meet operational targets. Special attention has been directed towards making sure that
the proposed solutions are feasible to implement in the specific department. One constraint has
therefore been that proposed methods must not have the potential to interfere nor deteriorate the
quality of the care given.
The theoretical purpose of this thesis has been to generalise the case study findings from the
radiology department to other ambulatory care service providers in emergency hospitals. It has
also been to analyse the applicability and effects of utilising different production planning and
control measures, tools and principles when managing an ambulatory care service provider in an
emergency hospital. The thesis has also aimed to provide conclusions regarding the theoretical
boundaries and constraints on production planning and control methods in the setting of an
ambulatory care service provider in Sweden.
Our literature review, as can be seen in chapter 3, furthermore show that radiological healthcare
and ambulatory care service are not widely studied from a production planning and control
perspective. This thesis will therefore aim to fit in this gap in theoretical research and provide a
foundation for further theoretical studies.
1.2.1 CONTRIBUTIONS
This study is highly relevant both from the perspective of Danderyds sjukhus in particular but also
Swedish healthcare in general. A more efficient radiology department will have positive ripple
effects and aid the current process oriented improvement projects conducted at Danderyds sjukhus.
It is also useful for the general healthcare debate in Sweden since change work is put in place in
other hospitals as well to better cope with demand.
This thesis should be seen as an analytical study of production planning and control for an
ambulatory care service provider inside a major emergency hospital in Sweden. It is representative
for the healthcare setting but also more focused on ambulatory care services. The healthcare
organisation at Danderyds sjukhus is also an example of an organisation divided according to
specific competences in departments with, at times, inexplicit connections.
A radiology department, or an ambulatory care service provider, in a major hospital in Sweden
can be said to operate in an industry that is permeated by a number of factors that have implications
from a production planning and control perspective. This study can therefore be said to concern
production planning and control in an industry that is to a large extent driven by individual
specialists able to influence all aspects of the operations (Andersen et al., 2014; Clegg et al., 2013;
De Vries et al., 1999). The industry can also be seen as highly influenced by culture. Decision-
making can therefore at times be unclear and operational changes become contingent on successful
bottom-up implementation. It can also be said to be subject to the influence of a number of external
stakeholders, including elected officials. The industry is therefore exposed to top-down decision-
making that is influenced by public opinion. Ambulatory care service units are furthermore
4
required to follow and adhere to technological trends and continuous innovation that need to be
incorporated in the daily operations. The operations are also controlled and governed by key
performance indicators that at times have an implicit connection to the actual situation in the
industry.
1.3 RESEARCH QUESTIONS
In order to achieve the purpose of this study one main research question and three sub-questions
answering up to the main research question were formulated.
1.3.1 MAIN RESEARCH QUESTION
How can production planning and control aid and improve the work process of an ambulatory
care service provider in a major emergency hospital in Sweden?
1.3.2 SUB-QUESTIONS
1. How can the work process in an ambulatory care service provider in a major Swedish
emergency hospital be mapped out from a production planning and control perspective?
2. What general challenges and opportunities exist for an ambulatory care service provider
in a major emergency hospital in Sweden from a production planning and control
perspective?
3. What activities should an ambulatory care service provider in a major emergency hospital
in Sweden prioritise in order to improve the situation using techniques, methods and
principles of production planning and control?
1.4 DELIMITATIONS
Since this is foremost a case study of a single department in a single hospital, there will be a bias
towards that specific environment. For benchmarking purposes, only one other hospital was
studied; “Universitetssjukhuset i Linköping”, chosen for already having implemented new ways
of working with production planning and control.
The literature review has been focused on theories and frameworks that describe and analyse the
empirical setting. For this reason only techniques already used in healthcare was evaluated, due to
it being possible to compare findings with literature of these methods. Testing new techniques in
a live setting was not part of the scope of this thesis.
The employer and facilitator of this study has been the department head of the radiology
department at Danderyds sjukhus. The department head is responsible for all medical and
operational aspects of the operations in all modalities of the unit. The thesis has therefore focused
on medium and short term production planning and control, since that is the scope of the current
work within operations management in the department. Long term planning in the radiology
department at Danderyds sjukhus is highly contingent on the construction of a new hospital
building currently under construction. This inhibits the potential for long term planning with
today’s resources, and planning of capacity and systems design is therefore disregarded. Facility
layout techniques have also been disregarded, since the design of the new department is finalised,
but still under construction.
1.5 THESIS OUTLINE
This section describes the outline for the thesis. The aim is to give a brief, incomprehensive, review
of what is discussed in all chapters.
5
Introduction: The thesis begins with an introduction that gives a background to the study and
introduces the relevant theoretical area of production planning and control. The introductory
chapter also presents the problem formulation with research questions and delimitations of the
study.
Method: The method chapter chronicles the research methods used in the study. The qualitative
case study conducted at Danderyds sjukhus that forms the basis for the study is outlined in the
chapter. The main methods used are also described and analysed in terms of validity, reliability
and generalisability.
Literature and Theory: The literature review presents the current theory in relevant fields that
formed a basis for the subsequent analysis. Production planning and control in general is discussed.
This discussion is complemented with a review of relevant research in production planning and
control in healthcare and radiological healthcare in particular.
Empirical Setting: The chapter describing the empirical setting aims to provide the reader with the
background information needed to comprehend the rest of the thesis. The chapter gives an
introduction to Danderyds sjukhus in general and the radiology department in particular with data
on historical demand as well as a description of the structure of the department. The radiology
department at “Universitetsjukhuset i Linköping” which is used for benchmarking purposes is
introduced. An overview of the different modalities available in the radiology department at
Danderyds sjukhus is also included.
Results: The results chapter outlines the empirical results and findings of the study. The empirical
results revolve around the radiology department at Danderyds sjukhus and figures based on
interviews, data and observations gathered in this study are presented. The most relevant findings
from the benchmark study in the radiology department at “Universitetssjukhuset i Linköping” are
also presented.
Analysis: The findings are analysed using relevant theories on production planning and control
discussed in the literature review. The emphasis has been to interpret, apply, analyse and generalise
the empirical results using production planning and control theories and reasoning. The chapter
aims to put the empirical findings in a more general context.
Conclusions: The conclusions chapter presents the final conclusions of the thesis work. Answers
to the posed research questions are provided, suggestions for future research as well as a discussion
of the thesis from a sustainability perspective. Recommendations directed specifically to the
radiology department at Danderyds sjukhus are also discussed.
Bibliography: A complete list of the works cited is listed in alphabetical order.
6
2 METHOD In this chapter the research method used for conducting the study will be outlined, described and
explained. A qualitative case study at the radiology department at Danderyds sjukhus formed the
basis of the study. The main method used for empirical research was semi-structured interviews.
There was also some consultation of quantitative data. The chapter begins with an introduction to
the research design and the methodological approach. The primary data gathering methods are
discussed in the context of the study together with how the data was analysed. The chapter also
addresses the issues of generalisability, reliability and validity of the study.
2.1 RESEARCH DESIGN
The research has been conducted using an abductive research approach and general conclusions
based on findings in specific empirical cases are presented. Abductive reasoning is an iterative
approach that allows going back and forth between empirical data collection and building a
theoretical framework theory (Feilzer, 2010). The data collection and analysis of the thesis has
been adapted and adjusted according to both theoretical and empirical findings during interviews
and in the literature review. The iterative model has enabled an investigation of the actual problem
of interest in general, and not only at the studied department (Dubois and Gadde, 2002).
A case study been chosen as the primary empirical data source since case studies are appropriate
in situations where the researchers seek to understand special circumstances. This is also valid in
situations where the research requires extensive and in-depth description of some social
phenomenon (Yin, 2009). Case studies can be defined as research situations where it is infeasible
to investigate all interesting variables given the number of data points, as in the case of this study
(Gibbert et al., 2008; Yin, 1999).
The empirical study presented here is based on a case study at the radiology department of
Danderyds sjukhus that has included an investigation in both ways of working and operational as
well as strategic problems. The case study has formed the basis for generalised analysis of
production planning and control in the context of radiology departments in emergency hospitals.
Data gatheringProblematisationand Analysis
Empirical data gathering
Semi-structuredInterviews
Quantitativeproduction data
Benchmark study
Theoretical data gathering
Books
Reports
Conclusions
Interviews
Literature studies
Meetings
Figure 1: Research Design
7
The outcome of the whole study is mainly of a qualitative nature, where ideas of how to work are
presented. Findings of a qualitative nature from interviews have been treated with an emphasis on
source criticism and have, to the extent possible, been checked in relation to the literature review
and quantitative data made available from the hospital. The findings and the situation at Danderyds
sjukhus has also been analysed in relation to a benchmark study at the radiology department at
“Universitetssjukhuset i Linköping”.
2.1.1 PROBLEM IDENTIFICATION AND DEFINITION
The thesis project was initiated by the department head of the radiology department at Danderyds
sjukhus. In order to map the situation as well as to identify issues to study at the radiology
department and the hospital in general a series of semi-structured interviews was conducted as part
of a pre-study. The interviewees were recommended by the department head and interviews were
held with managerial staff on both a department and hospital wide level. Some findings from these
interviews were used in the empirical data collection but problem identification was the main focus
at the time. Another purpose with the initial interviews was to ask for appropriate subsequent
interviewees, and in doing so, increase the reliability and validity of the study.
A preliminary literature review, further outlined below, was also conducted in parallel with the
interviews. This was done to be able to put the preliminary findings in a theoretical perspective
and lay a foundation for the rest of the thesis.
All the interviews were recorded, with approval from the interviewees, and thorough notes where
taken together with a discussion afterwards between the students regarding key takeaways. The
interviews were approximately 30-60 minutes long and performed by both researchers.
2.1.2 LITERATURE REVIEW
The empirical data gathering was preceded by a thorough literature review aimed at forming a
preliminary theoretical framework. The literature review was performed by mainly reading journal
articles, reports from official Swedish organisations and to some extent books on interesting
subjects for the study. The search for articles was mostly done using Google Scholar but also KTH
Primo. From the beginning the terms used was “Production planning review”, “Production
planning techniques” and “Production planning healthcare”. The idea was to start with reading
review articles on production planning and control in order to get a broader sense of common
concepts and models but also to find additional sources. References found in read articles were
investigated and new searches were conducted if new relevant topics were found. The primary
focus was to find articles published in 2011 or later to the extent possible.
The literature review was then iterated in parallel to the empirical study. From there this thesis
could be positioned in a research context. This was aimed at ensuring the relevance of the study
from a more generalised perspective (Denscombe, 2004; Dubois and Gadde, 2002) .
A systematic approach is advisable for qualitative social sciences, which is what this subject is
touching upon (Denscombe, 2004). The development and progress of the literature review was
noted systematically throughout the process. This will make it easier for other researchers to follow
the work and therefore increase reliability (Gibbert et al., 2008), a topic that is discussed more
thoroughly in a separate chapter.
8
2.1.3 EMPIRICAL DATA GATHERING
The sources for empirical primary data in this thesis were primarily from Danderyds sjukhus but
also a field study at “Universitetssjukhuset i Linköping”. The primary empirical data gathering
method used for the study was a semi-structured interview.
2.1.3.1 Interviews
Semi-structured interviews are interviews that have a loosely pre-prepared transcript with open-
ended questions about the area to be researched. However, follow-up questions to answers given
are allowed which may take the interview in a different direction than first anticipated. This would
enable the exploration of unforeseen interesting details (Britten, 1995; Collis and Hussey, 2014).
The reason for conducting interviews was furthermore because they made it possible to obtain a
picture of what an interviewee thinks, does, or feels and to enable him or her to he or she was able
to give thorough answers to all of the questions (Collis and Hussey, 2014), which is what was
needed to answer the purpose of this study.
A series of 21 formal semi-structured interviews with people with expertise in different areas of
radiological healthcare were performed, mainly at Danderyds sjukhus but also
“Universitetssjukhuset i Linköping”. Interviews were held with: physicians, radiological nurses,
administrative staff, and the medical staff responsible for different modalities (i.e. CT, MRI etc.).
This ensured that the whole process would be assessed collectively by the interviewees. The
interviewees were promised anonymity. Interviews were also held with representatives for other
departments in the hospital, including the emergency and the orthopaedic department. The purpose
with these interviews was to gain a deeper understanding of the role of the radiology department
in the overall patient flow at the hospital, but also to identify problems.
The interviews were conducted with both the researchers present. This gave a better chance to pick
up on nuances and to ask better follow-up questions. It enabled the researchers to do a more
efficient information gathering to achieve valid findings. A downside with both being present is
that this may make interviewees feeling threatened and reluctant to speak freely at times (Collis
and Hussey, 2014). All interviews were however aimed at making the interview subject feel
comfortable. Furthermore, all the interviews have, with permission of the interviewee, been
recorded.
2.1.3.2 Production Data
The in-depth study at Danderyds sjukhus was further underpinned by internal quantitative
historical production data. The quantitative data consisted of data on examinations performed
historically, as well as fulfilment of performance objectives set up for the department. The analysis
of the data was assessed and sanity checked by representatives from the department. The findings
Table 1: Table of Roles Interviewed in Danderyds sjukhus and at “Universitetssjukhuset i Linköping”
9
were primarily used to understand the empirical setting and thus serve as a foundation for further
both theoretical and empirical analysis.
2.1.3.3 Benchmark Study
A benchmarking field study in the comparable hospital “Universitetssjukhuset i Linköping” was
performed with the aim to develop a deeper understanding of a problems radiology department
face on a more general basis in Sweden.
The reason for going to Linköping in particular was because they presented new ways of working
in the radiology department during the conference Röntgenveckan 2014. The benchmarking study
was conducted after the initial study at Danderyds sjukhus to avoid bias in the problematisation
and further data gathering at Danderyds sjukhus.
2.1.3.4 Observations
Observations of the nature and specifics of the operations were obtained by both researchers. These
were acquired by being present in the radiology department at different times and while being
shown the department by radiology nurses.
2.2 ANALYSIS OF EMPIRICAL DATA
Both quantitative and qualitative empirical data was gathered and analysed. The qualitative data
has been central to the problematisation and the analysis. All data gathered was analysed iteratively
in light of other information gathered, and a special emphasis was to analyse the empirical data
using theory found in the literature review.
The main goal with the quantitative data was both to challenge and underpin statements in
interviews and analysis as well as provide a deeper understanding of the setting from a production
planning and control perspective.
Since multiple sources for the analysis were used; qualitative data gathered in interviews,
quantitative production data together with observations of the work at the department and a field
study as a complement, to describe the situation, triangulation was utilised. This increased both
validity and reliability of the study (Collis and Hussey, 2014), two concepts discussed further in a
separate paragraph below.
2.2.1 QUALITATIVE DATA
The qualitative data gathered in interviews served both descriptive and delimitating purposes. One
main purpose was to produce a framework for the setting, in terms of e.g. constraints from a
production planning and control perspective. All suggestions and opinions unravelled were
analysed from the perspective of our own reasoning together with theoretical findings.
Qualitative content analysis can be defined as “a research method for the subjective interpretation
of the content of text data through the systematic classification process of coding and identifying
themes or patterns” (Hsieh and Shannon, 2005). A content analysis, with the aim to find patterns
in answers, was performed. The content analysis of choice was a conventional one since it fits best
with the study’s aim of developing new theories regarding how to work in a Swedish radiology
department (Hsieh and Shannon, 2005).
10
2.2.2 QUANTITATIVE DATA
Quantitative data was gathered at Danderyds sjukhus; it exported from the business intelligence
platform QlikView. The aim was to analyse and describe the nature of the operations from a
demand and performance perspective.
The quantitative data was analysed in different ways depending on the data. The performance of
the department was analysed based on time series over goal fulfilment of the department’s own
stipulated performance goals. This way it was possible to indicate areas of interest in the process
for further analysis.
Understanding the demand for radiology was analysed by plotting highs and lows in demand. The
data was visualised and analysed as time series over different time spans which enabled assessing
of demand fluctuations for different modalities.
2.2.3 GENERALISABILITY, RELIABILITY AND VALIDITY
Generalisability, reliability and validity have been considered as important concepts to keep in
mind and refer back to in social sciences research. Generalisability refers to the extent to which
research findings and conclusions from a research study can be said to be applied in a more general
setting outside the scope of the study. Validity is the extent to which the study represents what it
aims to measure. Reliability refers to the extent to which a study is replicable; if another researcher
does the same study, he or she should obtain the same results (Collis and Hussey, 2014). Errors
relating to randomness in the empirical data should thus be low in a reliable study.
Reliability may be difficult to assess when a case study is conducted based on semi-structured
interviews. No two interviews are the same since mood, pronunciation of questions, the follow-up
questions etc. are unique to the situation (Collis and Hussey, 2014). Therefore the interviews
aimed to be thorough when it comes to taking notes of e.g. what was done and what kind of follow-
up questions were asked. This will make it easier for someone to replicate the study (Gibbert et
al., 2008).
Reducing error relating to randomness was addressed mainly through interview saturation,
whereby interviews were held until further interviews provided no new information. Interview
saturation suggests that the study is comprehensible in the sense that no new information will be
unveiled through conducting more interviews (Francis et al., 2010; Gibbert et al., 2008). All
interviewees were also asked to refer to other interviewees that might shed different light on the
subjects discussed. Employees in the radiology department at Danderyds sjukhus furthermore
rotate between modalities and could therefore provide insights relating to all modalities of the
department. This enabled swift interview saturation, and the number of interviews held was
deemed appropriate to map different perspectives to a sufficient extent.
The case study utilised both qualitative and quantitative data findings to increase both the validity
and the reliability. Qualitative data findings are often associated with a high degree of validity, and
quantitative data with findings with a high degree of reliability (Collis and Hussey, 2014).
The validity of the study should be considered as high given that the study is largely based on
qualitative data sources (Collis and Hussey, 2014). Case studies are commonly exposed to dangers
of subjectivity that will decrease validity. The analysis also took into account that large portions
of the interviews focused on subjective perceptions and opinions relating to the subject. The
analysis aimed to describe the phenomena from employees’ perspectives this was necessary to
understand all issues and problems.
11
The validity of all measures and parameters studied were therefore be a major focus in the
preparatory phases. Steps was taken to ensure “a clear chain of evidence” throughout the report to
make it possible for readers to see if the study has stayed valid to its focus (Gibbert et al., 2008).
By beginning with a focus on literature and Danderyds sjukhus as a specific case, it was possible
to get a good understanding of the concept which made it possible to achieve a high validity for
the study. If the concept was not well understood at a theoretical level, nor at the client company,
it would have been difficult to contribute in a theoretical and empirical sense.
Generalisability is an interesting point when it comes to case studies. Historically, the focus on a
single setting has been regarded as a problem (Dubois and Gadde, 2002) since generalisability
implies that results are true in other settings (Gibbert et al., 2008). This stems from an old
positivistic view on research where generalisability is connected to scientific generalisation
(Dubois and Gadde, 2002). This is line with the natural sciences aim; to generalise empirical data
to a population (Gibbert et al., 2008). However, generalisation may also be viewed and explained
by analytical generalisation, which instead to generalises empirical findings to theory; not
population (Gibbert et al., 2008).
This thesis has aimed for analytical generalisability in order to ensure applicability of the outcomes
in general discourse. Case studies enable researches to put the conclusions in a bigger setting for
generalisation (Gibbert et al., 2008). The case study at Danderyds sjukhus was therefore deemed
suitable for development of new theories from empirical findings (Eisenhardt, 1989).
The case study at Danderyds sjukhus was conducted with aid from the department head of the
radiology department. The department head assisted in scheduling interviews with important
people for the process and gather data for the following analysis. It should however be noted that
the empirical data gathering was based on the requests by the authors by the report and the data
gathering was therefore not biased by any possible preconceptions of the department head.
All in all, the study should be considered to have high reliability, validity and generalisability, for
the reasons mentioned above.
12
3 LITERATURE AND THEORY The aim of this chapter is to present the current theory in relevant fields that formed a basis for
the subsequent analysis. The main fields discussed and described are production planning and
control and production planning and control in healthcare.
Theory has constituted the basis for the interpretation of results and formed an initial framework
in the scoping phase of the entire thesis. To gain a thorough understanding of theories behind
production planning and control, which constitutes the main theoretical field for the thesis, and
how it could be utilised in an ambulatory care service provider, the approach has been to begin
with general theory and move on to more specific themes. This has been important both as part of
the process of deepening the researchers understanding of the subjects as well as making it possible
to give a description of the current research in the field of interest. The fields of interest can be
seen in the main research areas listed below. The literature review will follow in subsequent
sections of this chapter and follow the path from more general to case specific.
Production planning and control can be defined as the planning of processes for manufacturing
and production in a company. It involves resource allocation of employee activities, materials,
production capacity in order to satisfy a demand that could be forecasted over a certain time at a
reasonable cost (Gelders and Van Wassenhove, 1981). Production planning deals with organising
and designing processes as well as deciding in advance who, how and when something should be
done. It deals with the activities in an operation that matches supply with demand. Control refers
to the continuous control needed for planned operations to be well-functioning (Slack et al., 2010).
For this thesis, as previously stated, focus will be on production planning and control of an
operation why the sections will be as follows:
Production planning and control: In this section the aim is to create an understanding of the field
production planning and control, how it has evolved, and what the current trends in the research
are. While the point is to create a broader understanding of the field, the scope will still be
somewhat limited to what has been deemed directly interesting to the context of this thesis. What
is presented in this section constitutes the basis of the thesis’ theoretical framework. The areas
presented in this section are:
Production planning and control
Development of production planning and control theory
Performance Requirements
Systems and networks perspectives in production planning and control
Production information flow
Performance measurement
Lean
Lean Healthcare
Production planning and control in healthcare: This section has aimed to describe production
planning and control theories in the context of healthcare due to it being the broad system in which
the case study is conducted. The review aims to describe the specifics, what is unique with
healthcare as opposed to manufacturing industries when it comes to production planning and
control.
13
Every environment comes with a unique set of aspects, and this study has had a special emphasis
on considering the challenges of healthcare in general and ambulatory care services in particular.
Given that the case study concerns radiological healthcare we have chosen to focus on that here.
Research efforts to put production planning and control in the context of radiological healthcare
have therefore been reviewed.
3.1 PRODUCTION PLANNING AND CONTROL
In literature, other terms are also to describe roughly the same phenomenon using different terms,
e.g. production planning and control (Hulshof et al., 2012; Slack et al., 2010), production planning
and control, PPC (Buzacott et al., 2012), operations production planning and control (Olhager,
2013) and manufacturing production planning and control, MPC (Jonsson and Mattsson, 2003).
Van den Berg describes planning activities; e.g. inventory management and storage location
assignment, affect short and medium whereas control activities pertains to operational decisions
affecting short term: e.g. routing, sequencing, scheduling and order-batching (Van Den Berg,
1999). Some describe the evolution and scope of production planning and control and thus
implicitly which activities constitute production planning and control. These activities typically
coincide with the categorisation by Slack et al. (Buzacott et al., 2012; Jonsson and Mattsson, 2003;
Olhager, 2013).
We have chosen to use the categorisation by Slack et al. since it coincides well with a seemingly
rather strict academic consensus as to the scope of production planning and control as well as what
activities constitute production planning and control. The main activities then that Slack et al uses
are loading, sequencing, scheduling, and monitoring and control:
Loading: Loading deals with how much to do in a workplace and could for example be the
decision on how much a machine should be working. In some industries loading could be
planned and decided beforehand, this is called finite loading. Meanwhile, in an emergency
hospital it is more difficult to limit the load, there one has to cope with the flow of demand
as good as possible instead which is called infinite loading.
Sequencing: Sequencing regards the order that work is conducted. Different constraints put
on an operation create a necessity for planning the sequence of work conducted.
Sequencing could for example be the prioritisation of patients in an emergency department.
A process that will be further elaborated upon in the results chapter.
Scheduling: Scheduling deals with planning of when something is supposed to be done,
e.g. time tables for airlines’ flights. Staffing is also part of scheduling, which is a complex
part of scheduling. This is especially a problem for healthcare where one needs to plan for
a set competencies required to be available in an environment of uncertain demand and
emergency cases.
Monitoring and control: Monitoring and control involves monitoring and controlling the
operation. It is indeed rather self-explanatory with its name, after loading, sequencing and
scheduling of the operation it needs to be controlled and monitored.
3.1.1 DEVELOPMENT OF PRODUCTION PLANNING AND CONTROL THEORY
Theoretical research in production planning and control has during the last century developed a
number of techniques and approaches towards improving and dealing with production planning
and control in organisations. This has been done in close connection with the empirical cases in
manufacturing since manufacturing companies always strives to be on the forefront (Olhager,
2013). The origins of production planning and control is in planning on a shop-floor level planning
14
where labour division and time studies were common in manufacturing plants (Olhager, 2013;
Taylor Frederick, 1911). The discipline of production planning and control has however expanded
and steadily become a more integral part of operations management in recent years (Slack et al.,
2010; Tonelli et al., 2013).
The introduction of computers during the 1950s played an important role in the development of
production planning and control, facilitating computer based systems to automatically monitor and
control an operation (Olhager, 2013; Zijm, 2000). Information technology enabled a new way to
automate and monitor production which gained traction. Demand and forecasting data could be
utilised in statistical models to optimise inventory and resource utilisation. Methods were on a
general basis mostly developed to adapt and optimise operations depending on the given
environment, and its conditions, that the organisation was acting in. The notion of dependent
demand was also challenged in the same era, which could be described as demand not needing
forecasting. Instead, it could be derived from higher-level items (Olhager, 2013).
The focus during the 1960s was then placed on inventory control since more companies attained
sizes that enabled a wide range of new possibilities. For this reason, research focused on efficiency
in handling large volumes (Umble et al., 2003). The theory on how to optimise capacity
management was widely debated over the same time period, but the main contention to focus on
resources that constrain capacity, and in turn productivity, was popularised by Goldratt et al. in the
book “The Goal” (Olhager, 2013). Theories on constraints can also be described from a number
of perspectives and have been so by many academic researchers (Goldratt, 1990). Gupta and Boyd
stress that constraints are factors that production need to be adapted to (Gupta and Boyd, 2008).
During the 1980s a paradigm shift in how to improve efficiency and effectiveness of operations
took place, especially with the introduction of the concepts Just-In-Time (JIT) as well as optimised
production technology/theory of constraints (Conti et al., 2006). In the early 90’s the scope of
planning expanded to include for example human resources, communication systems, and project
management. The term enterprise resource planning; ERP, was coined to describe this planning
(Umble et al., 2003).
The popularity of the JIT-concept came with the flooding of the European and American market
by cheap and qualitative products produced in Japan. Manufacturers in Europe and America were
forced to look to the Japanese practices of productions (Radnor et al., 2012). JIT could be clearly
be distinguished from the, at the time prevailing, Material Requirement Planning (MRP) systems
in the sense that it relied on organisational changes, to only deliver products when needed, rather
than computerised planning systems (Olhager, 2013). These systems were criticised for including
several assumptions that give implications when they are used a representation of the real world
(Zijm, 2000).
The Toyota Production System (TPS) was introduced with the JIT and Kanban system, with the
promise to dramatically reduce setup times, lot sizes, and lead times, by Sugimori et al. (1977).
The Toyota Production System subsequently revolutionised vehicle manufacturing and later
influenced other industries (Mazzocato et al., 2010).
This influence came in the Western world when TPS was later developed into the term Lean in
1988, and was being globally popularised by Womack and Jones with their book “The machine
that changed the world” released in 1990 (Ahlstrom, 2004). It has been called the biggest
revolution of operations management in the last 50 years (Slack et al., 2010). The Lean concept
has grown in popularity and become one of the most influential manufacturing paradigms in recent
15
times (Holweg, 2007). From the beginning most of the research focus on Lean was put on
manufacturing, but later researchers found interest in applying the Lean concept on service
operations such as healthcare as well (Ahlstrom, 2004).
The need for accurate and reasonable production planning and control systems has emerged from
the understanding that many manufacturing systems do, in fact, operate in a stochastic
environment. One way to face demand is by flexibility in the operation, either by model flexibility
aiming to meet demand with what is popular at a certain time or flexibility in the production
(Georgiadis and Michaloudis, 2012).
The reliability of all production planning and control models have therefore been gradually debated
in recent years, due to their representation of demand and other external factors. Assumptions of
for example a perfectly predictable demand are rarely fulfilled by reality, but managers were rather
advised to change the manufacturing system than adapt the models to fit the situations (Zijm,
2000). This notion has affected the development since the choice of how to face demand naturally
has a significant impact on the operations design (Tonelli et al., 2013).
From the development outlined above, it could be derived that the scope of the techniques utilised
in production planning and control has grown with time, covering more areas of the operation.
From a start inventory control was the main focus but has moved towards production schedules
for an operation together with other firm activities such as sales (Olhager, 2013). Companies have
begun to act on the belief that it is not possible to compete successfully in a global market based
on internal efficiency alone. The focus of competition has changed from between companies to
between supply chains (Christopher, 2012). van der Meer-Kooistra and Vosselman note that the
boundaries of firms are continuously subject to change (Van der Meer-Kooistra and Vosselman,
2000).
Nguyen describes a trend where production planning and control activities are often required to
assume a holistic perspective for the organisation as a whole. The planning environment is often
required to be aligned and match the frameworks employed in order to realise positive results from
implementing new planning methods (Nguyen et al., 2014). This goes in line with other theories
stating that institutional, strategic, cultural and historical factors are of great importance to control
structures in an organisation (Van der Meer-Kooistra and Vosselman, 2000).
The discussion around applicable frameworks for describing the current situation seems to agree
on the need to go beyond the boundaries of the organisation. Discrepancies exist in how to go
about modelling this reality. How one views the organisation does indeed affect which production
planning and control methods, concepts, and instruments are applicable (Müller, 2006). These can
be grouped in different ways, but in this study it is reasonable to categorise them as either online
or offline planning (Hans et al., 2012). Offline planning are more passive approaches conducted
in advance, entailing to optimise variables in accordance with a systems dynamics analysis. This
approach takes the surroundings as given, and adapts operations instead. Online planning does
however consider the environment as interchangeable and focus on how it can be affected by the
current company. In this approach one can for example view all actors involved in the process as
a network.
16
3.1.2 PERFORMANCE REQUIREMENTS
Flexibility of production is an area with growing interest in manufacturing plants today. The aim
is to be more responsive to meet a competitive market (Jimenez et al., 2014). Today customers
have grown accustomed to rapid changes via e.g. mobile apps (Akkermans and Van Wassenhove,
2013). The term mass customisation has gained traction among manufacturers and service
providers further enlightening this trend. Customers want to have products available according to
their specifications, preferably anywhere in the world (Mourtzis and Doukas, 2014). This put
further pressure on production planning and control of these production plans since flexibility need
to be in mind to a great extent (Akkermans and Van Wassenhove, 2013).
To achieve sustainable success and adhering to operational objectives it is important to use relevant
performance measurements that track the performance of the organisation. Performance
measurements can be used to overview an operation, to keep track of implementations and to spot
deviancies from the plan. Performance measurements forms a link between planning, action and
results (Micheli and Mari, 2014).
Good performance measurement goes beyond financial data; it is important to measure multiple
things like quality and customer satisfaction as well (Eccles, 1990). The focus of the measures
needs to be on key strategies of the corporations. This is what usually becomes problematic for
many organisations; failure to translate qualitative targets into quantitative metrics. The
importance of succeeding with this is further stressed by the fact that intangible assets are
becoming gradually more important and tend to be more difficult to quantify (Bhasin, 2008). All
in all performance measurements can be categorised in two areas of usage. One is the part of
implementing change to, tracking and improving an operation. Performance measurements can
also be useful tools for communication, both internally and externally (Melnyk et al., 2014;
Micheli and Mari, 2014).
3.1.3 SYSTEMS AND NETWORKS PERSPECTIVES IN PRODUCTION PLANNING AND CONTROL
Theory has turned towards relaying and understanding the dynamic and stochastic nature of
production in PPC systems with a focus on flexibility and adapting to the surroundings of an
organisation. This focus perhaps stem from the need to identify and remedy the discrepancies
between what has been planned and actually performed activities dynamically (Georgiadis and
Michaloudis, 2012). This literature study has identified two main perspectives for understanding
an organisation’s external environment: a networks perspective or a systems dynamics perspective.
The systems perspective involves modelling of what could happen in a system of actors and how
to deal with it. In the literature review on systems dynamics and discrete event simulation in
production planning and control, the major focus has been found to be on handling different
scenarios and optimising production in accordance with predictable scenarios or events, rather
than anticipating everything (Georgiadis and Michaloudis, 2012; Hall et al., 2006; Martinez‐Moyano and Richardson, 2013; Nguyen et al., 2014; Tako and Robinson, 2012). Academics seem
to agree on that it is more important to focus on what could happen rather than when or if it does,
especially in an environment with stochastic demand. Georgiadis and Michaloudis adopt a system
dynamics approach to study the behaviour of a complex manufacturing system’s response to
different demand types and special events. Their investigation reveals that it should be possible to
optimise production based on historical data and still be relatively invariant to real-time events
(Georgiadis and Michaloudis, 2012).
17
As previously discussed flexibility has gained traction with shorter product lifecycles and
increased changes in demand. It has also been discussed that there is an increase of focus on supply
chains and the operational process outside of the internal operation. This facilitate possibilities for
systems dynamics to be used more to model the environment and can thus be used as a foundation
for long term operational planning (Vlachos et al., 2007).
Tako and Robinson applies systems dynamics and discrete event simulation in a logistics and
supply chain context and conclude (Tako and Robinson, 2012). Martinez-Moyano has reviewed
system dynamics literature to determine best practice in system dynamics modelling. Their study
reveals that there is no general consensus regarding how to model system dynamics (Martinez‐Moyano and Richardson, 2013). However, we still believe that the underlying assumptions and
purposes of the approach may be important to adopt in the healthcare setting. Martinez‐Moyano
and Richardson advocate an iterative system dynamics approach process that highlight the key
parts of the process, and has the overall goal to understand problems and the system (Martinez‐Moyano and Richardson, 2013).
Viewing an organisation as part of a supply chain network is another approach to understanding
the external implications from a production planning and control perspective. Müller explores the
notion of considering manufacturing as a collection of competence-cell-based networks. He states
that the key organisational form of the 21st century are autonomous elementary units of production
operating in collaboration in temporary networks, where non-hierarchical collaboration is a
prerequisite (Müller, 2006). Others argue that competitive advantages in firms are largely driven
by knowledge transfer and collaboration (Argote and Ingram, 2000). Not working towards the
same goal, and optimising on objectives particular for different parts of the supply chain is
undesired and labelled sub-optimisation (Goddard et al., 2004).
3.1.4 PRODUCTION INFORMATION FLOW
Internal communication is gaining in importance as an organisational function, and has developed
to an independent research field (Verčič et al., 2012). This study has for this reason delimited the
theoretical review from internal communication theories and instead focused on communication
surrounding demand and productivity inside of the operation.
Lean and JIT are examples of pull systems. In order to achieve an internal pull system it is common
to utilise so called Kanban cards, that dictate what to produce next for different production units
(Zijm, 2000). It is a communication tool authorising movement through the process making
delivery to the next step at the right time attainable which facilitates a smooth flow (Joosten et al.,
2009; Persona et al., 2008). Referrals in hospitals can be seen as examples of Kanban cards
(Ahlstrom, 2004). The utilisation of Kanban cards can be seen as a way to provide visual means
of flow management in a process (Turner et al., 2012). A problem with Kanban is if it is used
wrong systematically in an operation, work-in-process inventories may increase. This is an issue
that has been identified in a cyclical demand environment (Ebrahimpour and Modarress Fathi,
1985).
In MRP and ERP systems, on the other hand, one could argue that a push system is assumed where
the company sets up plans and communicate productivity goals top down (Zijm, 2000). The
literature study would suggest that these approaches towards production planning and control of
productivity are becoming rare in large organisations.
Turner et al. uses a systems engineering approach in a software development context and concludes
that the usage of Kanban cards maximises the system value flow (Turner et al., 2012). Other studies
18
also stress the potential importance and effects of optimised utilisation of Kanban in manufacturing
environments (Ciemnoczolowski and Bozer, 2013; Köchel and Nieländer, 2002). Goddard et al.
further stress the importance of information flow as a mean to control an operation, highlighting
its need for being discussed (Goddard et al., 2004).
An important factor for communication in an organisation is learning and the communication of
knowledge. Learning in an organisation can be said to go through three phases – experience
accumulation, knowledge articulation and knowledge codification. In the first phase the learning
is communicated by doing and using. In the second phase the knowledge becomes articulated and
the learning can be reflected upon and thought about. In an organisation characterised by
knowledge codification the learning is written down, adapted and implemented in the activities. It
is transferred from local experts to codified manuals bringing benefits for the organisation. While
experience accumulation may be positive from a job specialisation perspective, codification can
bring more to an organisation, when ideas are written down they become available to more people
who can analyse and develop them (Prencipe and Tell, 2001).
3.1.5 LEAN
As discussed earlier, the term Lean comes from Womack and Jones in a book on the Toyota
Production System. JIT and Kanban are examples of two techniques closely connected to Lean
but the term in itself is larger than that as will be discussed here. From the automotive industry it
permeated several industries and became a prevailing management philosophy (Olhager, 2013). It
is also the management philosophy that the Danderyds sjukhus state that they strive to use.
The main idea of Lean is to have an operation that aims to reduce waste (non-value adding
activities), to always strive to enhance productivity and to create an organisational culture of
continuous improvement. This culture should involve everyone in a firm, not just the management
level. Generally, Lean is achieved by using different tools and techniques developed to reach the
attributes of Lean stated here (Ahlstrom, 2004; Radnor et al., 2012).
Lean is not only a set of tools, however, it is also widely accepted that it is a way of thinking that
should permeate a whole organisation with a Lean way of thinking in everything that is done
through the whole operation including external factors such as the supply chain. Implementation
has been debated in research, namely why different results are yielded from different
implementation cases studied. In healthcare there has been a big focus on tools during an
implementation rather than a system-wide approach. The same happened in the automotive
industry in the West first turned to the Japanese production ideas. This may be a reason why
implementations sometimes fail, but it is not applicable to call it a certainty with the research
available. The failure is usually in the creation of an organisation devoted to continuous
improvements (Radnor et al., 2012).
Having a smooth flow is also a large part of Lean production and a way to eliminate waste. This
is further stressed by the importance the concept just-in-time has gained, it is closely connected to
Lean by emphasising that things should only be produced when needed. Then be delivered to the
next step of a process just in time for when the next step needs it. This is the description of a pull
system (Ahlstrom, 2004; Slack et al., 2010).
Other important concepts are: involvement of everyone (to be able to achieve continuous
improvement), zero defects by focus on quality in production, a concept closely connected to
eliminate waste since zero defects implies no waste. Lean also has an impact on workforce level
with multifunctional teams and decentralisation of responsibilities. Teams should be responsible
19
for several tasks within an area, helping workers realise their full potential (Ahlstrom, 2004). This
marks a step away from the more granular labour division once prevalent in manufacturing plants
across the Western world (Olhager, 2013).
The role of the worker have otherwise been a debated topic, namely that Lean leads to stress and
some have dubbed it “mean production” focusing on for example poka-yoke (a concept that
products should be fool proof for the next part of the process) which eliminates creativity and a
higher production tempo. When Lean is implemented in a good fashion it has been shown that
Lean does not lead to increased job stress (Conti et al., 2006). It could be argued that the
multifunctional teams and involvement of everyone help workers enjoy their jobs more which
could reduce stress (Ahlstrom, 2004; Conti et al., 2006).
Something that is important to have in mind is that all the concepts are highly dependent on a good
knowledge of both the internal and external processes of the operation, since Lean spans the entire
operation (Ahlstrom, 2004). A way to spread the culture needed, to involve everyone and make
sure that that the organisation strives to go in the same direction is for each team to have daily
stand-up meetings, called so due to them being short and not require sitting down. The idea is just
to have a quick briefing of the current situation (Middleton and Joyce, 2012).
Regarding the terminology around the concept Lean, several variants to describe roughly the same
thing are prevalent in the literature. Sometimes it could be called just Lean, but terms as Lean
production, Lean synchronisation and Lean thinking are also used to the production ideas based
on the Toyota Production System. This indicates that there are several views on what Lean actually
is, which is problematic when discussing the topic (Ahlstrom, 2004).
3.1.6 LEAN HEALTHCARE
After the rise of Lean for production, the idea of using Lean in a healthcare environment evolved.
Today a concept called Lean healthcare has been introduced. Its origin is not exactly known, but
the idea to use Lean in healthcare dates back to 2001 (De Souza, 2009). This could be seen as a
natural evolution of management research since the healthcare setting fits well for Lean theories.
Waste could mean the difference between life and death in a hospital. A smooth flow is something
that patients, and thus healthcare personnel, enjoy and it facilitates the possibility of giving care to
more patients using the same amount of resources. Some use of just-in-time also comes natural in
a hospital, since it resembles the natural setting (Ahlstrom, 2004).
The healthcare environment puts a certain pressure on the implementation of Lean since it has its
own set of environmental variables. Olaitan has discussed and described troubles of implementing
Lean in such a multiple product environment, especially regarding JIT. This is due to the fact that
production capacity needs to be shared between different products (Olaitan et al., 2014).
There are several successful examples of Lean implementation in healthcare (Bowerman and
Fillingham, 2007; De Souza, 2009). Improvements have been noted in different areas e.g. time-
savings, productivity enhancements, reduction of errors, improved staff satisfaction and reduced
mortality (Mazzocato et al., 2010). Some of the findings are debated though since a main idea of
Lean is that everyone should be involved and therefore requires an organisational shift. The
implementations discussed in research tend to be done in a few departments rather than hospital
wide. This will only provide weak support to the idea of the Lean hospital which would require a
hospital wide implementation (Andersen et al., 2014; Burgess and Radnor, 2013). Some research
findings point toward the need for data and thorough training in Lean thinking among personnel
to be able to successfully implement Lean (Andersen et al., 2014). An area where a successful
20
implementation has shown improvement outside of efficiency is the stressful environment
healthcare can pose (Bowerman and Fillingham, 2007).
As discussed in the previous section the focus has been on tools in Lean implementation for
healthcare. A further problem is one of healthcare in general which will be discussed more in the
subsequent section; implementations have often been in certain departments and not the whole
hospital. This do not reflect the holistic view stressed by modern production planning in general
and Lean in particular (Mazzocato et al., 2010).
3.2 PRODUCTION PLANNING AND CONTROL IN HEALTHCARE
The call for utilisation of production planning and control in healthcare has arisen from the
introduction of new demands and the increasing commercialisation of healthcare systems
worldwide. In the last decades, since the 1970s, there has been a pressure on healthcare to become
more efficient due to cost pressure and increased demand. Concepts are commonly imported from
different manufacturing industries. There has also been an increased focus on the quality of care
that is given to a patient (Beliën and Demeulemeester, 2008; Brailsford and Vissers, 2011; Radnor
et al., 2012).
This has coincided with a general change of discourse regarding how public services should be
managed. The term New Public Management, NPM, was coined during the 80s and has been
discussed since in the Western world. It describes a collections of ideas regarding how public
organisations could import new ways of working in general to improve efficiency, be more like
private organisations and less bureaucratic (Dunleavy et al., 2006; Pollitt et al., 2007). The pressure
therefore has been on public organisations in general but since healthcare is a large part of public
spending, it is of great interest (Andersson, 2014).
3.2.1 HEALTHCARE FROM A PRODUCTION PLANNING AND CONTROL PERSPECTIVE
The healthcare setting is unique from a production planning and control perspective for a number
of reasons and governance is indeed multifaceted. Little research has been put into supporting the
applicability of different techniques in the healthcare setting. Another detail to keep in mind is that
when you try to apply operations research in the context of healthcare is that a special emphasis
need to be placed on implementation (Brailsford and Vissers, 2011; Nguyen et al., 2014). There
are a number of environmental variables or aspects that significantly impact the effectiveness and
sustainability of the methods employed (Jonsson and Mattsson, 2003).
3.2.1.1 Governing Dynamics
The healthcare setting is impacted by a wide range of stakeholders, as it is run by elected counties
and tax funded in several countries. This exposes hospitals to a long list of external factors that
impose constraints on operations. Most importantly, the common interest entails a high interest
from policy makers that can affect any and all aspects of the operations. The preconditions for
healthcare organisations are therefore highly contingent on the country in which it operates
(Brailsford and Vissers, 2011). This emphasises the need for objective and transparent decision-
making, and there has been a lot of focus put on this in academic research (Harper, 2002).
The governing dynamics of the hospital systems also put strains on capacity and resource
management, and require sophisticated models that take into account the inherent complexity of
the environment (Harper, 2002). This underlines the need for central production planning and
control systems on a hospital level, to satisfy the needs of patients and adapt capacity accordingly
(van Merode et al., 2004). Historically, this has been a problem for healthcare. The different
21
departments tend to be autonomous which lead to mangers only looking inside their own
department when working with production planning and control (Hans et al., 2012).
3.2.1.2 Specialist Driven Environment
In a range of European countries, hospital management cannot always assume full control over
output as it is performed by clinical specialists that have contract with other hospitals (Brailsford
and Vissers, 2011). Many hospital processes are driven by medical specialists who have the main
influence by their profession. However, they do not manage the process on paper (Clegg et al.,
2013; De Vries et al., 1999).
This specialist environment can become a hurdle when implementing changes to the operation
(Clegg et al., 2013). Thorough evidence that a new method is useful to the organisation is needed
to be able to convince personnel (Andersen et al., 2014). While each patient is unique in a sense
and medical treatment complex, it is possible to standardise hospital operations more by using
appropriate models. This may be a problem to implement though since it hampers a physician’s
freedom at work (Testi et al., 2007). There is a case to be made for more systems engineering
implementations in healthcare, letting the operation become more data-driven. This give less
influence to healthcare specialists which may lead to a clash at implementation which is important
for managers to have in mind (Kopach-Konrad et al., 2007). Research show that failed
implementations of e.g. knowledge management systems have been partly due to the strong
specialist culture that permeates healthcare organisations (Radnor et al., 2012).
The nature of the hospital organisation also inherently imposes constraints on the room for
manoeuvring in terms of production planning and control. Many hospitals are arranged in units
that contribute to the process with different competences and clinical specialities. This leads to
issues with the patient journey through the hospital, stemming from the fact that no one is in charge
of the customer process as a whole, and the different medical practitioners want to preserve their
autonomy. This, in combination with the aforementioned external stakeholders’ influence, results
in a complex command structure that is not always straightforward and different stakeholders need
to be managed (Brailsford and Vissers, 2011). The importance for competence cell networks was
discussed above for production planning and control in general (Müller, 2006), and it should be
noted that this is also prevalent in healthcare. There has been an increase of interdisciplinary teams
in the healthcare environment to meet the increased demand on specialisation (Kilgore and
Langford, 2009).
3.2.1.3 Resource Utilisation and Process Flow
Optimising the usage of resources in a healthcare setting involve coordinating all resources in the
production planning and control processes. When speaking of resources one typically in the
healthcare context refers to the different human resources, the medical equipment and the facilities
(Blake and Carter, 2002; Vos et al., 2007).
Process flow has been identified as central to the efficiency in a hospital environment and patient
volume needs to be considered especially in relation to limited resources (De Vries et al., 1999).
Patient flow is considered similar to process throughput in traditional manufacturing industries.
Process bottlenecks in clinical and administrative tasks can delay patient treatments and
discharges, thus leading to lower quality and higher costs. Problems with the flow may for example
lead to patients waiting too long in the hospital and exposing them to infections, a unique aspect
of healthcare (Devaraj et al., 2013).
22
When designing a hospital layout, it is important to have the healthcare strategy in mind, but also
take the future and uncertainties into account (Vos et al., 2007). From an operations management
point of view, hospitals need to maximise throughput using the resources at hand, while also taking
into account different requirements for flexibility in delivery (non-emergent or emergent patients)
(Vos et al., 2007). One suggestion is to divide hospitals into different operational processes, e.g.
deterministic and non-deterministic ones1. This would facilitate a better flow through different
processes (van Merode et al., 2004).
3.2.1.4 The Patient
Another major difference when comparing the healthcare setting to traditional manufacturing
facilities is the patient; in healthcare he or she is both the customer and at the same time involved
in the production process (Hall et al., 2006). In general, the customer is also the one who decides
which product to acquire and is to an extent involved in the product development (Mourtzis and
Doukas, 2014), something that is only in the hands of policy makers and medical specialists in
healthcare, marking a significant difference.
Experiences from previous research show that the only person who sees the whole process of the
patient in a hospital is the patient itself. In general, each department only care for its own part of
the process (Bowerman and Fillingham, 2007). This rules out a standardisation of products and
processes due to the high variation between individuals, hospitals and even practitioners in the
same speciality (Brailsford and Vissers, 2011). The patient also disables many possibilities to
automate several activities, since the care provided needs to be centred around different individuals
(Zijm, 2000).
3.2.1.5 Defining Value
Healthcare systems are generally highly congested, which further inhibits the potential for
planning (Hall et al., 2006). Furthermore, since it is difficult to define value in this environment,
it becomes difficult to measure productivity and efficiency in a way that would satisfy all
requirements of a healthcare operator (Cameron et al., 2001). Defining demand becomes more
difficult at the same time since it includes both the number of patients and their well-being. The
needs are also varying among patients (Nguyen et al., 2014).
In public business, the interest for performance measurement has grown in parallel with the
introduction of NPM. Defining good targets in healthcare has been problematic since performance
measurement was imported from the business world, and publicly employed managers tend to be
unaccustomed to the thinking of the private sector (Arnaboldi and Azzone, 2010). Bad
measurements could have dire consequences for a business. If they are irrelevant people simply
will not care and the purpose of measuring is removed (Bhasin, 2008). However, if used well it
has proven to be a useful method for efficient management of an organisation (Melnyk et al., 2014;
Micheli and Mari, 2014).
3.2.1.6 Demand Uncertainty
Demand uncertainty poses another issue for hospital production planning and control, and it has
proven difficult to generalise frameworks for forecasting and adapting to demand. Hospitals with
emergency departments are also required to handle interruptions in the form of for example
unexpected arrival of patients and the fact that patients’ conditions may change during the
treatment process (Hall et al., 2006). Indeed, demand can be considered to be almost random in a
1 Examples of this could be to make a distinction between emergency and non-emergency patients. The work process
in examining a non-emergency patient can be said to be deterministic since all contingencies and individual
preferences of non-emergency patients can be known beforehand.
23
general sense (Womack and Jones, 2010). These variations in demand creates extra workload
pressure, especially for nurses (Beliën and Demeulemeester, 2008).
Before patients can be admitted they also require an initial screening to determine profile
complexity and medical urgency, in order to enable decisions concerning the treatment process
(Vissers et al., 2001; Vos et al., 2007). All emergency procedures put pressure on flexibility in all
hospital operations as to cope with the variability of demand (De Vries et al., 1999). It has been
shown to be of operational benefit with better matching of nurse and physician schedules (Beliën
and Demeulemeester, 2008).
3.2.2 FRAMEWORKS OF PRODUCTION PLANNING AND CONTROL UTILISED IN
HEALTHCARE
From the above stated aspects of healthcare operations it can be deduced that production planning
and control initiatives are often aimed at being put into different frameworks for analysis. These
naturally aim to distinguish and scope so as to not overcomplicate the analysis. Among these is
the prevalent framework that highlights hospitals from different managerial areas as it is divided
into different hierarchical decompositions (Hans et al., 2012).
There are some prevailing frameworks for hospital planning developed in the recent decades: Roth
and Van Dierdonck (1995), Vissers et al. (2001), and Hans et al. (2012) (Nguyen et al., 2014). It
should be noted however that all these frameworks make assumptions on environmental variables
that the hospital is acting in (Nguyen et al., 2014).
Roth and Van Dierdonck base their framework on the ideas of manufacturing systems approaches
and material requirements planning is central. The framework is built on requirements that liken
the hospital environment to mass production with extensive capabilities and capacities for
production planning and control (Nguyen et al., 2014; Roth and Dierdonck, 1995).
Vissers et al. formulate a framework which includes five levels of production planning and control:
1. patient production planning and control, 2. patient group production planning and control, 3.
resources production planning and control, 4. patient volumes production planning and 5. control
and strategic planning. The framework aims for coordination of these five levels with decision
models regarding resource management and patient flow (Vissers et al., 2001).
Hans et al. focus on efficient hospital decision making and encompass the entire healthcare
organisation. They conclude that the then existing framework they analysed, concerning hospital
production planning and control, was too narrow since it focused on a single managerial area and
disregarded others. They proposed a framework that should encompass all levels of operations as
well as hierarchical production planning and control (Hans et al., 2012).
3.2.3 PRODUCTION PLANNING AND CONTROL IN RADIOLOGICAL HEALTHCARE
This section outlines the most distinguished and relevant research found in our literature search on
the topic of production planning and control in ambulatory care service providers. After an initial
screening we chose to focus on radiological healthcare, since it is the empirical setting of the case
study, and the research on ambulatory care services in general was limited.
When comparing different departments of a hospital, the radiology department, as presented
earlier, could be classified as an ambulatory care service, defined as a department that provides
care but not a bed and room for the patient to stay in. All of the patients can be seen as outpatients
referred from other departments, where they after examination will be transferred back. Other
24
examples of ambulatory care services are primary care services, community services and some
hospital-based departments such as the endoscopy department or the outpatient clinic (Hulshof et
al., 2012).
Given this setting, it is obvious that radiology cannot be planned as a stand-alone production
system, it should rather be planned as a part of a larger system (Blackmore, 2007). The goal of the
planning should be that the stay at the radiology department should be as short as possible, both to
achieve flow and to minimise the risk for patient complications occurring away from originating
care unit, thus being unable to provide appropriate treatment (Simpson and Falter, 2014).
It is important to remember that planning in radiological healthcare is inhibited by the varying and
changing needs of patients, as in the case of all healthcare practice. This creates a high degree of
uncertainty in production planning and is a main issue to handle in ways that enable smooth flows
from a resource perspective (Nguyen et al., 2014).
A radiology department is dependent on a smooth flow and working equipment for efficiency
(Kruskal et al., 2012). To be able to achieve this smooth flow, theory suggests it is useful to look
towards Lean and Kanban. Due to the nature that radiology is composed of advanced machines
photographing body interiors, the department is often in the front when it comes to incorporating
new technologies in the hospital. Information technology (IT) also has got an important role to
play since the speed to produce images and storage for these are important aspects for efficiency
in the radiology department (Ozcan and Legg, 2014; Rao and Levin, 2011). With the technological
advancements in radiology the need for strategic planning increases (Ozcan and Legg, 2014).
Technologies here refer to different techniques for radiology, e.g. Computed Tomography (CT)
(further outlined in the empirical setting chapter). Investments in technologies lead to increased
costs applying further pressure on radiologists to cut costs in their departments when a hospital as
a whole wants to implement cost cutting procedures (Sunshine et al., 2010).
There are examples of how modern technology has had an impact on production planning in
radiology from a broader scope than just radiological techniques. Instead of having a specialised
radiology department other departments are empowered and educated to perform simpler
radiological examinations directly; e.g. the cardiology department, performs radiological
experiments to detect heart issues. Since the images are easily transferable, a radiologist could be
consulting from anywhere. This is however a disputed development among some radiologists since
the role also means the performing of the important task of deciding who should get examined first
in a situation with referred patients from different departments (Rao and Levin, 2011).
As picture quality improves as well as transmission speeds, the case for teleradiology, remote
radiologists assessing scans, has also become more appealing in different cases. This is especially
valid for smaller hospitals with a lower amount of total staff, which then can have the radiology
department online 24/7 without a radiologist available at a certain time. It is also possible to offer
more services in rural areas this way (Binkhuysen and Ranschaert, 2011), which means that present
radiologists do not get replaced by teleradiologists as has been feared. Instead, it opens up new
services. The fact that radiologists have fears regarding this development is in line with earlier
presented research pointing to the reluctance towards change among medical specialists (Andersen
et al., 2014).
Competent staff is, together with the availability of imaging equipment, the most important
resources to a radiology department. Both nurses and physicians tend to specialise in advance
imaging areas such as angiography and magnetic resonance imaging (MRI), creating a lower
25
flexibility among units with several specialists modalities (Ozcan and Legg, 2014). There have
been attempts to use simulation models for defining the perfect staff mix, a common procedure in
healthcare research. However, these usually represent a perfect world which may not always be
available (Wang et al., 2012). With technological development leading to more modalities, staffing
becomes more difficult since more specific competences are required (Choy and Novelline, 2013).
26
4 EMPIRICAL SETTING This chapter gives an introduction to Danderyds sjukhus in general and the radiology department
in particular with data on historical demand as well as a description of the structure of the
department. An overview of the different modalities is also included with the aim to provide the
reader with the background information needed to comprehend the rest of the thesis. Lastly, the
radiology department at “Universitetsjukhuset i Linköping” will also be introduced and be used
as a benchmark in the subsequent analysis.
4.1 DANDERYDS SJUKHUS
Danderyds sjukhus is one of the largest emergency hospitals in Sweden, employing more than
3700 people. It services approximately 440 000 inhabitants in the northern parts of Stockholm with
both emergency and specialised healthcare with a focus on the most common diseases. The
hospital is owned by the Stockholm County Council divided into 12 business areas including
emergency care, cardiac medicine, surgery and urology, internal medicine, orthopaedics and
radiology. Academic research is conducted in collaboration with Karolinska Institutet (Danderyds
sjukhus AB, 2015b).
Quality improvement work at Danderyds sjukhus is based on the Lean philosophies for healthcare
and aims to fulfil the medical care agreement and the definitions for good healthcare (“God vård”)
stipulated by the Swedish National Board of Health and Welfare. In 2013 Danderyds sjukhus as a
whole reached 90 percent of the quality goals stipulated by the Stockholm County Council, and
was remunerated accordingly. The strategic plan for the hospital consists of steering documents
specifying long term ambitions and goals. Hospital management yearly devises a strategic
scorecard with specific performance indicators for the coming year. This scorecard is subsequently
broken down for each scope of practice in the hospital.
The hospital began working towards implementing a process oriented governance system fulfilling
the requirements for ISO9001-certification in 2013 and a complete mapping of all process and
support processes is presently conducted. Driven partly by the desire to cope with the increased
demand seen in general, Danderyds sjukhus has entered a program that will aim to improve the
overall production planning of the healthcare services offered at the hospital on a general basis
(Danderyds sjukhus AB, 2015c; Jacobson et al., 2015).
The flow of emergency patients in Stockholm County council was analysed in 2013, which
included a mapping of the stages that patients go through at Danderyds sjukhus. The mapping
illustrates how radiology is central in the diagnosing phase for most patients.
Currently a new building is under construction at Danderyds sjukhus which will house the
radiology department in close proximity to the emergency department. The idea behind having a
closer proximity is to facilitate a closer cooperation between the different departments. This would
be beneficial since a majority of the radiology department’s patients are emergency patients. This
will also mean a slightly bigger radiology department in the future with more labs than today.
27
4.2 RADIOLOGY DEPARTMENT AT DANDERYDS SJUKHUS
The radiology department at Danderyds sjukhus is subordinated in the scope of practice “Bild och
Funktion” and performs radiological examinations on patients referred from other departments of
Danderyds sjukhus, but also from external clinics in Stockholm County. The department is
showing strong numbers and have been performing better than budget in the recent 3 years straight,
continuously lowering costs while increasing revenues2. The increased revenue comes from an
increased demand for services. As can be seen in Figure 2, the number of examinations has
followed a cumulative annual growth rate of 6% yearly since 2008.
Examinations are conducted at all hours during the day, but the flow varies between different
hours; and most examinations are conducted between 07 and 16 (as seen in Figure 3).
Figure 3: Examinations conducted at different times (percentages, 2014)
2 Internal documents
Figure 2: Number of examinations conducted in the radiology department at Danderyds sjukhus
28
The department distinguishes between patients referred for emergency examinations and non-
emergency patients referred for scheduled appointments. The emergency appointments are
performed year round and at all times in the day whereas non-emergency patients are, as can be
seen in Figure 4, typically scheduled between 06 and 19. Scheduled appointments account for
approximately 30 percent3 of all examinations (Danderyds sjukhus AB, 2015a).
Examinations are being conducted in the following different modalities:
Computed Tomography (CT)
Magnetic Resonance Imaging (MRI)
Computed radiography (CR)
Fluoroscopy (XA)
Angiography (AI)
Medical ultrasound (US).
Each modality can further be divided into multiple labs where the examination is conducted.
CT: 3 labs
MRI: 2 labs
CR: 4 labs
XA: 1 lab
AI: 1 lab
US: 4 labs
3 From interview with the department head
Figure 4: Emergency and scheduled examinations (percentages, 2014)
29
The number of patients visiting the different modalities differs greatly and also varies over the
year, as can be seen in Figure 5. The dip in July is mostly due to it being the holiday period.
4.2.1 MODALITIES IN THE RADIOLOGY DEPARTMENT AT DANDERYDS SJUKHUS
To give a further introduction to a radiology department, each modality at the department will
described briefly in terms for example technology, usage areas, and safety considerations.
4.2.1.1 Magnetic Resonance Imaging
Magnetic Resonance Imaging (MRI) is a scanning technique invented in 1971 that utilises strong
magnetic fields to investigate anatomy and physiology of the body. It does not use X-ray beams
as traditional radiology. MRI produces cross-sectional images of organs and internal structures.
The technique can distinguish between different tissues and substances within the body thanks to
different reactions to the magnetic field. Contrast agents may be utilised for some MRI
examinations (Center for Devices and Radiological Health, 2015a).
The U.S. Food and Drug Administration states the primary uses of MRI scans as (Center for
Devices and Radiological Health, 2015a):
Abnormalities of the brain and spinal cord
Tumours, cysts, and other abnormalities in various parts of the body
Injuries or abnormalities of the joints
Certain types of heart problems
Diseases of the liver and other abdominal organs
Causes of pelvic pain in women (e.g. fibroids, endometriosis)
Suspected uterine abnormalities in women undergoing evaluation for infertility
MRI examinations provide different information than examinations in other modalities, including
detailed images of ligaments and cartilage. MRI and CT are to an extent complementary, where
both technologies have their advantages and limitations worth mentioning. MRI is often
recommended as an alternative to other radiological techniques since the technique does not
Figure 5: Examinations conducted over the year for different modalities (percentages, 2014)
30
expose patients or physicians to ionising radiation. It also works better than CT for several
diagnoses, especially ones relating to soft tissues whereas CT is superior for hard tissue, such as
bone fractures (Center for Devices and Radiological Health, 2015a).
There are risks with MRI procedures that require special attention in clinical practice. All patients
need to be reviewed for contraindications, such as pacemakers and cochlear implants that may be
inhibited by the strong magnetic field, prior to MRI scanning. The strong magnetic field may also
cause accidents where objects are attracted to the magnetic field that cause a so called projectile
effect where the object is launched into the MRI machine. The magnetic field may also cause
undesired nerve stimulation. Another risk is for examinations to become intolerable for patients
with claustrophobia due to the fact that the patient needs to lie in the centre of the large magnet for
longer periods of time. Compared to CT, the examinations are rather time consuming (up to 40-60
minutes depending on the protocol for the MRI and 15-30 seconds for the CT). Overuse of MRI
has however been discussed from apprehensions relating to cost effectiveness and over-diagnosis.
For these reasons, MRI is often not recommended as the primary procedure for radiological
examinations (Center for Devices and Radiological Health, 2015a).
4.2.1.2 Computed Radiography
Computed Radiography (CR) is the classical radiography and is typically the first technique learnt
among practitioners. X-rays are produced and sent through a body where different tissues absorbs
different amount of radiation making it possible to produce an image by collecting the rays on the
other side of the body. Then, with the help of computers, a digital 2D-image is produced. This
process is quick, both when it comes to the examination and image processing. It is a matter of a
few minutes in total, as long as the image is correct and not has to be re-shot (Center for Devices
and Radiological Health, 2015e).
At Danderyds sjukhus, CR is primarily used to examine hard, primarily fractures in orthopaedic
patients, referred from the emergency department. It could also be used to examine lungs, different
cancers, scoliosis, infections etc. Generally, more advance radiological modalities, such as CT and
MRI, can produce better, 3D images providing for more secure diagnosis. But CR is often good
enough, and compared to other modalities, relatively quick and cheap, making it very useful as a
first radiological method where it is applicable. It is the first modality that nurses learn when they
begin to work at DS.
Since CR is an X-ray method, using ionising radiation, specialists should take measures to avoid
exposure for longer times. Computed radiography is comparatively inexpensive and simple. The
technique is adequate for many diagnosis questions but provide less information than more
advanced techniques such as CT and MR (Center for Devices and Radiological Health, 2015e).
4.2.1.3 Computed Tomography
Image tomography is an old technology from early 1900s where images are produced from several
splices of an object. CT is a more modern version of this where computers are used to produce
images out of several virtual splices that give a good picture of the inside of the body. As in CR,
X-ray beams could be used. However, in CT, computer processed X-rays are used to take several
2D pictures around an axis of rotation, making it possible to render a 3D view of the inside unlike
in CR where a 2D picture is the outcome. Compared to CR, better images providing more exact
renditions are produced making it suitable for diagnostic use in head, lungs, pulmonary angiogram,
cardiac, abdominal and pelvic and extremities. This come at the cost of a higher price, more
radiation than CR and a longer time to process images making CR an attractive option for more
simple examinations even if CT yields a result of higher image quality, that may be necessary to
31
answer some of the questions stated in referrals from departments (Center for Devices and
Radiological Health, 2015c).
4.2.1.4 Angiography
Angiography, or vascular radiology, is mainly used to visualise blood vessels and organs from the
inside; commonly veins, arteries and chambers of the heart. The technique is physiologically
interventional and involves injecting the patient with contrast agents in the blood vessels, most
commonly in the femoral artery, and imaging typically using fluoroscopy.
Angiography is an invasive medical procedure and may lead to complications on rare occasions.
These complications are both minor ones such as bleeding or bruising but major complications
such as blood clots, strokes, allergic reactions, kidney damage and cardiac arrhythmias do occur
(NHS Choices, 2015).
4.2.1.5 Medical Ultrasound
Ultrasound imaging was developed in the 1940s. The technique produces real-time images of the
inside of the human body using high frequency sound waves. The real-time images enable
physicians to investigate movement of body organs as well as flow through blood vessels.
The advantages of ultrasound compared to other radiological techniques are that it provides images
in real-time, is portable within a hospital, is relatively inexpensive and that it does not expose the
examined body to any ionising radiation. World Health Organization published a report in 1998
stating: “Diagnostic ultrasound is recognised as a safe, effective, and highly flexible imaging
modality capable of providing clinically relevant information about most parts of the body in a
rapid and cost-effective fashion" (World Health Organization, 1998).
Ultrasound examinations do however require special competence and is operator dependent. It is
widely recognised that a high level of skill and experience is required for high quality images that
facilitate diagnoses. The examinations are therefore, in Sweden, conducted by physicians (Center
for Devices and Radiological Health, 2015b).
The clinical practice of ultrasound has gone through a paradigm shift in recent years, starting in
the US, whereby ultrasound is used in emergency healthcare to a greater extent and not exclusively
in the radiology department. This evolution is further spurred on by the development of new
portable ultrasound machines that are compatible with for example smartphones and tablets (Cook
et al., 2007).
4.2.1.6 Fluoroscopy
This technology has several similarities with classic CR in its setup, with the big difference being
that a live film is produced from a constant flow of X-ray beams. Therefore it is very useful to
study ongoing processes inside the body like flow in blood vessels or as a guide when orthopedic
surgery is performed. Since a film is produced and flows are studied, the examination could be
comparatively long; up to 75 minutes. This means a big exposure to radiation making it important
for physicians to weigh benefits versus potential harm, severe burns is a rare side effect of the
examination (Center for Devices and Radiological Health, 2015d).
At Danderyds sjukhus this modality is mostly used to study processes in the gastrointestinal tract,
for example the swallowing process for patients with brain damages. This kind of examination is
usually quick, just a couple of minutes.
4.2.2 OPERATIONAL GOALS RADIOLOGY DEPARTMENT AT DANDERYDS SJUKHUS
32
Danderyds sjukhus is currently fulfilling 90 percent of the quality goals stipulated on a national
level for healthcare. The only stipulated goal pertaining to the radiology department is that a
patient, with a scheduled appointment at a certain time, has the right to demand the money back if
the examination is not done within two hours.
As can be seen in Figure 2, a major share of the examinations are emergency examinations. Since
the patients then are sent back to the emergency department the radiology department becomes an
important part of the national goal that a stay in the emergency department should last four hours
at a maximum. Because of this there is an internal goal that the maximum time from that the
physician refers a patient to a radiological examination to a preliminary diagnosis should be two
hours at a maximum for emergency examinations. Other than these operational goals the only
other operational monitoring is in the budget that should be met at year end.
4.2.3 OPERATIONAL PERFORMANCE
The goal fulfilment in the department can be said to follow a positive trend for the most part, but
have in recent years declined marginally. As can be seen in Figures 6, 7 and 8, the department fail
to fulfil the internally stated operational goal of two hours from referral to diagnosis for emergency
examinations. One of the reasons for this could be that the total volume of examinations is
increasing4.
4 For 2015 the share is only calculated on data for the first quarter
Figure 6: Number of emergency examinations with preliminary response within 2 hours after
incoming referral (2014)
33
Looking at goal fulfillment in general in Figure 7, we notice that the fraction of examinations
fulfilling the performance goals has been flat since 2008 without any signs of sustainable
improvement.
When it comes to waiting times in general, especially CT fails to meet the target of two hours.
This is without an increase in the number of examinations a seen in Figure 8. Members of staff
have indicated that the demand for emergency examinations is growing for CT. Non-emergency
patients then have to wait more since an emergent one may get prioritised ahead of the scheduled
appointments.
60 %66 %65 %
69 %70 %71 %67 %
61 %
0 %
10 %
20 %
30 %
40 %
50 %
60 %
70 %
80 %
90 %
100 %
2008 2009 2010 2011 2012 2013 2014 2015
Figure 7: Figure 7: Fraction of emergency examinations with preliminary response within 2 hours after incoming
referral
Figure 8: Fraction of examinations conducted within 2 hours after arrival
34
4.3 RADIOLOGY DEPARTMENT AT UNIVERSITETSSJUKHUSET I LINKÖPING
A field study was performed at the radiology department at Universitetssjukhuset i Linköping.
Compared to Danderyds Sjukhus it serves more patients, around 160,000 but a lower share of
emergency examinations, around 50%5. Research is also conducted at the department. Otherwise
it has roughly the same modalities as the radiology department at Danderyds sjukhus, with the
addition of a machine combining Positron Emission Tomography and CT (PET-CT):
CR
CT
MRI
PET-CT
Medical Ultrasound
Interventionistic (e.g. angiography)
It is the largest radiology department in the County Council of Östergötland. A further discussion
will follow in the benchmark analysis in the results section.
During the yearly conference Röntgenveckan (Radiology week) in 2014 the department head
presented a new way to plan and control the operations of the department which had been
beneficial for the department from several areas such as well-being of staff and production.
5 according to an approximation expressed by interviewee in benchmark study
35
5 RESULTS This chapter outlines the empirical results and findings of the study. The empirical results are
centred on the radiology department at Danderyds sjukhus. All figures presented in this chapter
are based on interviews, data and observations gathered in this study. The most relevant findings
from the benchmark study in the radiology department at “Universitetssjukhuset i Linköping” are
also discussed.
In this chapter the focus will be on the current situation from a production planning and control
perspective, the patient is not the focus point but highly involved as an actor in the process of the
radiology department. From the findings presented here, the process will be analysed in
accordance with current theories in the following chapter Analysis.
5.1 THE GENERAL PROCESS AT THE RADIOLOGY DEPARTMENT
Danderyds sjukhus aims to view the whole flow through its operation from a patient perspective
and the process outlined below can be fitted into the general process under “Investigation and
Diagnostics”, as seen in Figure 9.
The general process flow at the radiology department at Danderyds sjukhus can be seen as a
generic process with three phases: “Preparation”, “Examination”, and “Diagnosing and
Documentation”. As can be seen in Figure 9, the radiology department is only responsible for the
physical handling of patients during the “Examination” phase, but need to handle referrals in all
phases. During the interviews it was evident that this process was known to all radiology nurses
responsible for different modalities, but not documented as done here.
Every phase can, in turn, be divided into several steps. The steps look differently depending on
modality, but the overall process follows the three phases and the steps explained below.
Figure 9: The radiology process fitted into the journey of the patient
36
5.1.1 PREPARATION PHASE
The preparation phase concerns all activities after a patient is referred to the radiology department
up until that the patient arrives in the department.
Incoming referral: A referral is sent to specific modalities in the radiology department when a
physician in another department, at Danderyds sjukhus or an external healthcare provider, decides
to order a radiological consultation for diagnosing. The referral should include all relevant
information to the radiology department; including symptoms to diagnose as well as patient
history.
Referrals are typically dispatched in the internal IT-system, in radiology called RIS/PACS
(Radiology Information System / Picture Archiving and Communication System) and are either
for emergency or non-emergency examinations. This system is connected to a big documentation
system used for the whole hospital. A small amount of the referrals arrive in paper form, which
generally creates extra work for the staff at the radiology department. The risk for errors also
increases with this relatively unusual way of communication compared to the general process.
Prioritisation: Almost all referrals are read through and prioritised by physicians before patients
are being scheduled for appointments. Patients requiring more immediate care are prioritised
higher, and scheduled sooner than patients with lower priority. The physicians also make
methodological choices such as whether or not contrast agents are needed. The physician may also
choose to refer the patient to another modality, if more appropriate, judging on the information
given in the referral.
Exemptions are for extremities in CR and heart and lung examinations. These are prioritised
automatically by the IT-system and nurses can begin examinations without consulting physicians.
This automation facilitates a smoother flow since the nurses are not required to coordinate all
activities manually. What is important to have in mind is that auto-prioritisation makes the
Figure 10: Steps of the Preparation phase
37
referring physician responsible for patient care regarding radiation. This would otherwise have
been at the hands of the prioritising radiologist.
Appointment scheduling: Appointments are subsequently scheduled in accordance with the
previous prioritisation and available capacity in respective modality. The procedures typically
differ for emergency and non-emergency patients and tend to be handled separately by different
persons.
Patient summoning and pre-preparations: Non-emergency patients will receive a confirmation of
appointment (often written) including time and instructions regarding necessary preparations.
Scheduling and summoning of emergency patients from the emergency ward is done using a
special emergency ledger in the medical records software. This step also involves coordination
between different professional groups. This is due to that there sometimes is a need for certain
information in a medical journal, such as weight and kidney function in case of contrast
examinations, which should be available before the arrival of a patient.
5.1.2 EXAMINATION
The examination phase includes all activities from when a patient arrives in the radiology
department to when the patient is sent back to where he or she was before coming to the radiology
department.
Patient arrival: What happens to a patient at the arrival to the radiology department is contingent
on if it is an emergency or non-emergency patient. Emergency patients are sent to the reception
from the referring department to the reception where they are gathered to their respective modality.
Emergency patients can also be directly led to the examination room by the janitorial service
handling internal transportation at the hospital. Non-emergency patients arrive in the reception of
the radiology department and are subsequently sent or called in to respective examination or
preparation room.
Figure 11: Steps of the Examination phase
38
Examination in lab: The examinations are conducted by radiology nurses in MRI, CT and CR.
Physicians conduct examinations assisted by nurses in US, AI and Fluoroscopy. The time
consumption for each examination is highly contingent on a range of factors, including:
Modality – the nature of examinations differ a lot for different modalities. Naturally, this
impacts examination times.
Method choices – there are in total 54 different method choices for an examination,
including for example different methods to deploy contrast agents. Of course, all these
impact the examination times differently.
Patient sickness – a more severely ill patient is harder to handle if she/he arrives in a bed
or similar. This could mean heavy lifts for nurses when moving these to the radiology
machine which impacts the work environment negatively.
Organ examined – some organs are for example hard to see which requires the patients to
be put in difficult positions and takes more time.
Examiner’s competence – the competence of the radiology nurse or physician impacts the
efficiency with which the examination is conducted. This has an especially great impact in
modalities requiring a smooth flow such as CT and US.
Patient sent back: After examination, a patient is sent back to where they came from before the
examination, where the response will be handled. This typically results in patients being sent home
or back to the referring department.
39
5.1.3 DIAGNOSING AND DOCUMENTATION
The diagnosing and documentation part of the process is a step mainly managed by physicians and
to some extent secretaries; radiology nurses are not involved at all here. The step involves
producing reviews to the images produced in the examination phase, sending responses and
documenting the progress.
Primary read: The images produced in the examination phase are here being studied by a physician
that diagnoses the patient. The preliminary response to the questions posed in the referral is written
and sent to the referring department via the IT-system. The examinations are categorised
depending on type and result. The time consumption for diagnosis varies considerably between
patients; one patient could for have a simple fracture which could be diagnosed in a matter of a
few minutes (or perhaps even shorter). A cancer patient with follow-up examinations to monitor
the cancer progress on the other hand takes more time since old pictures need to be consulted and
it could be difficult to detect tumours if they are small.
Double read: After the primary read, a second, usually more experienced, physician studies the
images and audits the preliminary diagnose and primary read. Pathological finds are sometimes
discussed in daily meetings with physicians from referring departments.
Response dictated and documented: The physician performing the double read either dictates a
new response or confirms the first response for documentation. Secretaries document the dictated
response in the patient’s medical record. They do not contact the patient with the diagnosis, instead
that is done by the referring physician.
Figure 12: Steps of the Diagnosing and Documentation phase
40
5.2 CONTINGENCIES OF DIFFERENT MODALITIES
The generic process, outlined in Figures 10, 11 and 12, is for the most part adhered to in the
different modalities, the process does however differ in some aspects. This section elaborates on
the peculiarities of different modalities, and describes the aspects that differ from the generic
process.
This section will assume the reader has gained a basic understanding of general specifics and
contingencies of the different modalities of a radiology department, as described above.
5.2.1 MAGNETIC RESONANCE IMAGING (MRI)
Prioritisation: All emergency examinations need to be prioritised right away and scheduled within
one week. This is a longer time frame when compared to e.g. CR and CT where examinations can
be scheduled sooner since there are more labs available and examinations are shorter. Moreover,
it is seldom a question of life and death thus possible to extend this time frame.
Scheduling: An enrolled nurse is responsible for the scheduling of examinations. The scheduling
requires an understanding of the necessary examination protocols; a quick reference guide is
therefore available and used. The scheduling also needs to consider account that the two different
machines used for MRI examinations are different models with some different attributes. The
differences are seen both in patient loading and unloading as well as in the images produced. The
department has identified a potential to save time by specialising machines for different organs
during a certain day. This way it is possible to avoid unnecessary set up times that arise when an
examination differs from the previous is to be performed. On average nine non-emergency and
three emergency patients are scheduled per machine and working day.
Patient preparations: The patient preparations and safety precautions prior to the examinations are
extensive and vital for magnetic resonance imaging. A preparatory checklist with patient
preparations is therefore attached to all summons of patients: e.g. if the patient has a pacemaker,
Figure 13: Contingencies in the MRI modality
41
stents or previous surgeries including metal implants, that may be hazardous in close proximity to
the magnetic field in the machine. The safety procedures on site are extensive and all patients are
reviewed using a checklist to ensure safe examinations.
Patient arrival: Emergency patients are often inpatients and therefore often arrive in beds escorted
by the janitorial staff. This generally makes the handling of patients simple except for the lift that
has to be performed at one of the MRI machines when a patient is loaded.
Long examinations: Examinations held in the MRI machines are comparatively long, due to the
extensive amount of images produced in an examination. The radiology nurses need to be attentive
to ensure that the patient lays still, otherwise there is a risk for blurry/unusable pictures. These are
otherwise common errors which may increase total examination time for a patient since new
images are necessary.
Staffing: There are extensive competence requirements for the radiology nurses who work in this
modality. In order to be eligible to conduct examinations on the MRI machines radiology nurses
are required to complete an extensive competence development course. The MRI machines are
therefore not part of the regular staff rotation schedules. Five radiology nurses in total staff the two
machines. Physicians are available to assist if needed during examinations.
Capacity: The modality is scheduled at full capacity on weekdays and only active one night every
week after office hours. Emergency examinations need to be prioritised over non-emergency
examinations, and the waiting times are therefore increasing if the number of emergency
examinations increases. Approximately 30 patients are forwarded to other clinics due to capacity
constraints at the radiology department at Danderyds sjukhus. More physicians and radiology
nurses are required to satisfy the identified demand for MRI examinations at the department, given
that the machines are available more than they are utilised.
Claustrophobic patients: Some patients get claustrophobic during the examination since the
patient lays in a tight, enclosed tube. The radiology nurses therefore often try to talk to the patient
in a pre-emptive manner to make them more comfortable and the patients are given a signalling
device to halt the examination if the claustrophobia becomes unbearable. If necessary, patients
could be given sedatives or even be anesthetised.
Collaboration with other departments: The collaboration with other departments works well in
general. Patients do however sometime arrive for emergency examinations uninformed about the
process and where they will receive their diagnosis due to lack of information from the referring
department. This is generally the case when a patient gets a referral for an MRI examination the
next day when visiting the emergency department. This leads to patients contacting the radiology
department with questions, using up nurses’ time. The interviewees suggest that better
collaboration and understanding between departments could eliminate this.
42
5.2.2 COMPUTED RADIOGRAPHY (CR)
Prioritisation: Examinations of extremities are prioritised automatically by the IT-system and
directly handed to radiology nurses for examination. In the normal manual process, the
prioritisation of computed radiology examinations are sometimes inhibited by referrals being
unsatisfactorily written, lacking information which leads to extra work for nurses who sometimes
themselves need to find out what to examine. Some referrals also ask for multiple examinations
with no clear diagnosis question posed, resulting in unnecessary activities, extra radiation for the
patient and a more difficult prioritisation process. Moreover, since it is contingent on comparison
between referrals. These problems could be the result of the high rotation among physicians in the
emergency department leading to many unexperienced physicians writing referrals.
Scheduling: There are four different types of referrals in total for the CR modality: emergency,
non-emergency, drop-in and hip process. Drop-in (patients with a referral are eligible to come
when they want) and hip process (emergency patients with a hip fracture) are special, relatively
low volume on a day to day basis but need to be considered when production is planned. The
patients classified as non-emergency are the only ones who are booked. About 70 percent of all
examinations in the modality are emergency examinations.
Patient preparations: Examinations generally requires no time consuming preparations, except for
that the organ to be examined need to not be covered by clothes or such.
Patient arrival: Scheduled patients need to be examined within 30 minutes; the patients are
otherwise reimbursed their fee. In the hip process, the patient arrives directly in a prepared lab to
speed things up. Regarding drop in patients the department have experienced some problems, these
patients tend to arrive on similar times, and are therefore forced to wait longer compared to if they
would have been scheduled.
Examination: One major inhibitor of the flow in this modality is the difficulties in producing
quality images for diagnosis. The efficiency of the flow is therefore highly contingent on the
Figure 14: Contingencies in the CR modality
43
competence of the examiners, since it will reduce the amount of insufficient images. Referrals
lacking specific information further lead to wasteful activities in the form of unnecessary
examinations.
Staffing: New employees always begin with CR and the modality is therefore staffed with a
majority of new employees. It is regarded as the most basic form of radiology. During
examinations each lab only requires a single nurse.
Capacity: Labs in the modality are open year round, therefore the major constraint on capacity is
set by the number of machines available for examinations. Another constraint is the potential
overuse stemming from referrals being written unsatisfactorily as discussed earlier.
Collaboration with other departments: The operational collaboration with other departments is
limited to referrals and responses. CR is the modality doing the most examinations, most of them
quick and routine and referring physicians do not except more than a quick diagnosis.
Collaboration is important regarding referrals since there has not been any discussion on how these
should be written which may create difficulties. Scheduling at the modality has become simplified
since access to the emergency ledger, used by the emergency department, was given to the
radiology department.
5.2.3 COMPUTED TOMOGRAPHY (CT)
Prioritisation: A large amount of both emergency and non-emergency patients are being
prioritised daily. Of the activity in the modality, emergency examinations account for
approximately 80 percent of the examinations and these need to be prioritised right away.
Figure 15: Contingencies in the CT modality
44
Scheduling: The scheduling is handled in two separate schedules: emergency and non-emergency
(one nurse responsible for each). The aim is to have a well thought out planning strategy, which
requires both operational consideration and clinical competence. The scheduling also involves
managing the waiting list for referring departments up to one year in the future. Some examinations
require updated results from lab tests, which may create problems with the division of
responsibilities such a setup entails.
Patient preparations: Preparation instructions are attached in the appointment confirmation sent
out to scheduled patients, still everyone is not ready at arrival. Patients scheduled for examinations
including contrast agents that is administered orally need to be appointed for arrival 2 hours before
lab appointment and require updated lab screen (an adequate serum creatinine level to determine
kidney function). The creatinine level must be below a certain value for the kidneys to be able to
handle the contrast agent without failure.
Patient arrival: Some contrast agents need to be administered well in advance before examinations
requiring instant setup at arrival. Other contrast agents are infused intravenously during the
examination, which requires the patient to sit and wait for some time as the contrast agents are
deployed. Jewelleries that might interfere with the radiation, and the radiological modality, need
to be removed before the examination.
Examination: The examinations are comparatively short and are typically booked in slots of 20
minutes. Examination times do however vary and can be as short as 10 minutes. Many different
protocols for examinations are used, and new protocols are rapidly developed. CT entails high
levels of radiation and radiology nurses therefore need to make sure to not produce unnecessary
images. The operational efficiency when conducting examinations differs a lot for different
persons, contingent on e.g. mobility of patients which may affect the loading of a patient and the
time needed for this.
Staffing: Two nurses staff each machine when the modality is working at full capacity. The
division of labour between these two nurses is clear; one is responsible for patient preparations
and the activities inside the lab, and the other one is responsible for controlling the machine
according to protocol. The environment is described as stressful with a high tempo, probably
aggravated by short examinations when the modality is working at full capacity.
Capacity: The modality is operating at full capacity; i.e. booking all available time slots day and
night year round. The increased amount of emergency appointments results in longer waiting times
for non-emergency patients.
Collaboration with other departments: There is not that much collaboration with other
departments. The referrals tend to be of high quality but some information can be missing, which
interrupts the flow of patients through the modality. A commonly occurring error in the referrals
is the requirement of recently acquired creatinine levels which are needed if contrast is to be
infused in the patient. Scheduling at the modality has become simplified since access to the
emergency ledger was provided.
45
5.2.4 MEDICAL ULTRASOUND (US)
Prioritisation: Some examinations are prioritised by radiology nurses without consulting a
physician, as opposed to the normal procedure where physicians prioritise all referrals. This is only
done for a few types of common and simple examinations, such as deep vein thrombosis.
Scheduling: Today, the nurses are considering the flow when scheduling for the modality, which
requires a competent nurse in charge of planning. The aim is to never schedule four long
examinations at the same time, since this might result in unnecessary build-up of queues. Instead,
longer examinations are scheduled in a specific lab so that the other labs may conduct several short
examinations at the same time. The modality is more contingent on staffing of physicians since
they perform the examinations and not the nurses as in the case for the other modalities. Non-
emergency patients are scheduled to account for half of the modality’s capacity. Some
examinations that could not be performed a certain day are rescheduled for early next morning
before the flow of emergency patients ramps up, which is unique for the US. Some examinations
are coordinated with the cytology department and the most intensive examinations are planned
extensively.
Patient arrival: The patient arrival follows the general process with no major difference.
Patient preparations: Needles are prepared for examinations where these are needed for injections
of contrast agent. Usually, some clean-up activities after previous patient must also be performed.
There are also examinations that require more extensive preparations. Additionally, the amount of
preparatory activities needed after patient arrival at the modality also varies depending on how
well prepared the patients are before they arrive.
Examination: Doctors conduct the examinations that are relatively short (around 10-30 minutes).
Sometimes it is possible for physicians to manage multiple patients simultaneously, since nurses
perform activities such as contrast injections during certain examinations. The examinations
Figure 16: Contingencies in the US modality
46
involve no radiation and limited lifts of patients which make the working environment
comparatively comfortable for radiology nurses. A major concern for the nurses is to ensure a
continuous flow in the different labs, as well as supporting physicians on a standby basis.
Staffing: Apart from the physicians who conduct the examinations, enrolled nurses are also staffed
in the modality for performing of simpler tasks.
Capacity: The modality only accepts patients from Danderyds sjukhus due to capacity constraints,
the waiting list is a month long for non-emergency scheduling. A tendency towards unnecessary6
referral of patients to the modality has been identified. This is believed to be due to the quality
measure for the emergency department, whereby a patient needs to be treated within 4 hours.
Ultrasound is a safe choice without radiation and it could provide a quick diagnosis. This may
make physicians willing to refer patients for ultrasound examinations even if other modalities
actually are preferable.
Collaboration with other departments: The collaboration with other departments has been
identified to work best with departments that have a nurse responsible for inter-department
coordination. Since ultrasound is a modality with a rather high emergency flow, the
implementation of the emergency ledger, has improved on the collaborative environment for this
modality as well. An extended utilisation of the scheduling applications in the medical records
system has further improved on the situation, most specifically the communication with the
surgical department.
5.2.5 ANGIOGRAPHY (AI)
Prioritisation: Prioritisation is seldom performed. Mostly non-emergency examinations are
conducted and the flow of emergency patients is limited, thus creating less need for prioritisation.
6 Referral of patients where pathological finds can be deemed to be unlikely upon closer investigation
Figure 17: Contingencies in the AI modality
47
Scheduling: The scheduling is contingent on physician availability; currently only one of the
employed physicians fulfils the competence requirements to perform all procedures possible at the
modality. Some days during the week are reserved for different types of examinations: Thursdays,
the lab is utilised by an ablation team working with tumour removal and on Fridays, only
emergency examinations are performed.
Patient preparations: Patient preparations are of great importance in this modality. In effect, the
department need to prepare the patient for an invasive procedure. The patients therefore need to
have fasted before the examination and the nurses also need to ensure a sterile environment. These
preparations are disruptive and time consuming to perform on site, and patients are therefore told
to prepare for appointments by e.g. shower in a specific way. The examinations involve breaching
an artery and nurses need to make sure that that the referring department is equipped with
compression bandages and a space to rest in bed afterwards. The collaboration with other
department is therefore natural and usually well-functioning.
Patient arrival: Patients in this modality are always inpatients and always in bed (need to lie down
after intervention into the artery). It is also important to make sure patients have fasted prior to
examination, because in case of emergency the patient may need to become intubated (breathing
tube down the larynx) and stomach content may end up in the lungs, thus limiting the respiratory
capacity.
Examination: Differs from other modalities in that the examinations are often therapeutic which
requires both a physician and a nurse to be present in the lab during the procedure. Radiation levels
need to be considered in the lab to a larger extent since the physician and radiology nurse are
present during the examinations. Protective gear is always worn.
Staffing: Competence requirements takes a long time to acquire due to the limited flow, therefore
only a few relatively common emergency procedures are taught to all physicians. This is to be able
to conduct vital emergency procedures at all times.
Capacity: The angiography lab is currently not working at full capacity in the sense that the
machines are not utilised full time, even during weekdays. The waiting time for non-emergency
patients is at the most 4-5 weeks (normally within 2 weeks).
Collaboration with other departments: Since patients are always inpatients and need to be sent
back to referring department for rest after examination, collaboration is necessary and it works
well today.
48
5.2.6 FLUOROSCOPY (XA)
Prioritisation: The nurse responsible for the modality receives a phone call for an emergency
examination and then contacts a physician for prioritisation. Non-emergency patients account for
most of the referrals to the modality. The flow is relatively small, each day one emergency and six
non-emergency patients are booked.
Scheduling: Needs to be coordinated with physician staffing, central scheduling is responsible for
this at the modality. When scheduling patients, it is necessary to consider logistics and for example
schedule with janitors in advance in order for the patients not to be late for the appointment.
Patient preparations: The preparations tend to be rather extensive compared to other modalities.
A thorough clean-up of the lab is necessary and usually a comparatively large amount of materials
are required for examinations, which must be prepared by the nurse.
Patient arrival: The patient arrival process for the modality is similar to the generic process, there
are special cases though where contrast agents are administered at patient arrival.
Examination: Doctors are responsible for conducting the examination while being assisted by a
radiology nurse. Safety considerations are important since radiation exposure is high due to a
comparatively long exposure to radiological beams. The examinations are still relatively short
compared to other modalities such as MRI and US. Often patients with multiple diagnoses are
referred to this modality, which may result in increased requirements in terms of care; e.g. tubes
and IVs that need to be managed by the nurse.
Staffing: One specially trained nurse is responsible for the lab, though it is not fully staffed all day.
If there is no scheduled examination in the modality, the nurse moves around to assist in CR and
CT depending on demand.
Figure 18: Contingencies in the XA modality
49
Capacity: The modality is currently working with excess capacity. This grants the potential to
conduct some CR examinations since the machine in the lab is capable of this.
Collaboration with other departments: The collaboration generally works well, the contact with
other departments is close which facilitates good collaboration.
5.3 BENCHMARK STUDY
As presented earlier, a benchmark study was conducted at “Universitetssjukhuset i Linköping”
with the purpose to understand the initiative of new tools for production planning and control at
their radiology department, which they have been working on since 2012.
Both departments are similar and different in a number of aspects. They have roughly the same
modalities and the general three-phase process that can be seen above is also true for the radiology
department at “Universitetssjukhuset i Linköping”. This would suggest that the departments
should be able to adopt similar solutions to issues encountered in the clinic7.
There is however a major difference in the external environment of the two departments from a
production planning and control perspective, namely the ratio of emergency and non-emergency
patients. The radiology department at “Universitetssjukhuset i Linköping” have a larger portion of
patients referred for scheduled appointments compared to the radiology department at Danderyds
sjukhus. Since non-emergency patients generally are connected to less uncertainty, this group
naturally lends itself better to production planning and control.
The radiology nurses in the radiology department at “Universitetssjukhuset i Linköping” are
divided in sections based on special competences pertaining to different organs rather than
modalities. The department furthermore employs a competence management system for all
modalities based on an illustration of layers of specialist competence. Each modality has an inner
layer of specialists; typically one nurse responsible for the modality who can perform all types of
examinations in the modality. In the next layer the department have a number of radiology nurses
able to perform most examinations. In the outer layer of a modality’s staff are nurses that are able
to perform the most basic examinations and are therefore able to staff the machines after office
hours.
The department in Linköping has furthermore devised internal central quality management and
performance measurements top down. Quality management is nonetheless to greatest extent
possible aimed at being managed by radiology nurses and physicians rather than management
continuously in the day-to-day operations.
7 The radiology department at “Universitetssjukhuset i Linköping” does however perform mammographic
examinations; a modality that the radiology department at Danderyds sjukhus has dismantled.
50
5.3.1 PRODUCTION PLANNING AND CONTROL SYSTEM IN THE RADIOLOGY DEPARTMENT
AT “UNIVERSITETSSJUKHUSET I LINKÖPING”
The initiative to work more with production planning and control was initiated when the waiting
time for non-emergency patients were deemed too long by the department’s operations manager.
On top of that the operations manager also felt that the overview of the operations was insufficient
and there were no targets that were followed up during the year making it difficult to monitor the
progress and improve.
To be able to improve, a target stating that 70% of the non-emergency examinations should be
completed within two weeks of examinations and a target that a certain amount of non-emergency
examinations should be done in each modality every year. With these well-defined targets an IT-
system was developed that monitor this on a day to day level. The nurse responsible for each
modality then imports how much that they can plan to do each day with the current working
schedule into the system. Then it is monitored how many examinations are done each day, if they
meet their own plan and how this reflects upon the larger target that a certain number of
examinations should be completed each year.
The instant feedback available in the software makes it possible to monitor the performance
compared to target live and if they lag behind in a certain modality they know where to allocate
more resources. The idea is that data, and not individual experiences, should lead the production
planning.
Implementation started in 2012 and it is still and ongoing project. Resistance and scepticism
toward change was said to be shown both among physicians and radiology nurses. Scepticism
towards modern management theories was prevalent and the question of why they could not
continue do as they always had done was raised. Managing a cultural change therefore became a
major issue in the implementation process, and special care was taken to anchor the changes.
Today, the staff express satisfaction with the new ways of working and state that they are less
stressed.
51
6 ANALYSIS In this chapter the empirical findings are discussed in the context of relevant theories on
production planning presented in the literature and theory chapter. The aim has been to interpret,
apply, analyse and generalise the empirical results using production planning and control theories
and reasoning. The reasoning in this chapter forms the basis for the conclusions presented in
chapter 7.
An ambulatory care service provider should be seen as a specialised environment from a
production planning and control perspective. A radiology department in a major hospital in
Sweden can be considered representative for a hospital bound ambulatory care service provider. It
is indeed a healthcare service provider that only handles patients on an outpatient basis referred
from other healthcare providers and is therefore also highly dependent on its surroundings.
This chapter describes, discuss and analyses the most relevant and interesting empirical and
theoretical findings from a production planning and control perspective. The first Section (6.1)
analyses the feasibility and applicability of production planning and control techniques, methods,
and mindsets in an ambulatory care service provider in Sweden. The applicability of production
planning and control in an ambulatory care service provider is illustrated by applying central
frameworks on the process devised in the case study. In Section 6.2 we discuss the typical flows
apparent in an ambulatory care service provider in a major emergency hospital in Sweden by
applying production planning and control reasoning to different flows and scenarios. Section 6.3
addresses different aspects of collaboration and coordination concerning an ambulatory care
service provider; including both department internal and interdepartmental collaboration and
coordination.
6.1 PRODUCTION PLANNING AND CONTROL IN AN AMBULATORY CARE SERVICE
PROVIDER
Assessing the feasibility, applicability and relevance of applying production planning and control
in an ambulatory care service provider have been one of the main purposes behind this study. We
analyse whether it is advisable and feasible for an ambulatory care service provider to implement
production planning and control in the daily operations. This is illustrated by applying the most
central framework for production planning and control to the case study performed under the scope
of this thesis.
6.1.1 DEMAND FOR PRODUCTION PLANNING AND CONTROL IN AMBULATORY CARE
SERVICE PROVIDERS
We have seen, after reviewing literature on operations management, as well as production planning
and control, that the healthcare setting is a good example of how organisations are viewed today.
The environment is complex and involves a wide range of competences, products and services.
Healthcare organisations are also examples of systems with a lot of different facets that need to be
managed and understood as a whole in order to facilitate successful planning. The focus on holistic
approaches; and avoidance of sub optimisation, that is apparent in many studies on production
planning and control in general (Nguyen et al., 2014; Zijm, 2000), and also the focus on systems
dynamics and network theories (Müller, 2006; Tako and Robinson, 2012), should therefore be
applicable in the healthcare setting.
An important finding in this paper is that a radiology department should be seen as a representative
ambulatory care service provider since the characteristics and challenges it faces are likely similar
to other ambulatory care service providers. Ambulatory care service providers are only involved
52
in parts of the patient’s physical journey through a hospital. As such, an ambulatory care service
provider is highly dependent on its surroundings, and the production planning and control is
dependent on external factors. An ambulatory care service provider is however likely involved in
a symbiotic relationship with surrounding departments in a hospital where there is a clear mutual
exchange where departments are dependent on each other. The effects of public policy that regulate
for example inpatient care units also affect an ambulatory care service provider indirectly through
its referring departments. For this reason it is difficult to plan an ambulatory care without regard
to its surroundings.
We therefore conclude that an ambulatory care service provider in a major Swedish hospital indeed
needs to focus on adapting and responding to its environment. However, in order to do this
efficiently however, the department need to understand its surroundings; both the needs and
actions of referring departments and impact of changes in public policy. It will also have to
promote flexibility of operations to enable swift adaptation to changes in its surrounding
environment. This goes well in line with the trend we have observed in academic research on
production planning in general where the objectives described as well as the scope of production
planning and control has evolved in current research (Akkermans and Van Wassenhove, 2013;
Jimenez et al., 2014).
The objectives of operations, and in turn production planning and control, in academic research
was previously for example quality, cost, speed, or flexibility. The trend however seems to be an
increasing focus on flexibility of operations (Jimenez et al., 2014). Organisations need to focus on
staying relevant by evolving to meet future demands as well as create new market places (Umble
et al., 2003). Hospital organisations however need to meet up to changes in the dictated demands
top-down.
The scope of production planning and control has furthermore expanded to including larger parts
of the supply chain, and to both model and control this environment from within the organisation
(Olhager, 2013). The need to understand the context that a particular organisation is acting in is
vital in order to facilitate production planning and control (Nguyen et al., 2014). Given the setting
that an ambulatory care service provider is active in we conclude that production planning and
control theories are focusing on solving similar problems. The focus on flexibility in production
planning and control stems from the progression towards faster pace regarding life cycles of
products in many industries (Vlachos et al., 2007).
As discussed in the literature review the understanding of ones surroundings can be done in
different ways, where a systems dynamics and discrete event simulation approach can be
contrasted to a network modelling approach: Discrete event simulation focuses on the simulation
of different events that might occur, and systems dynamic on modelling and how the situation for
the organisation will evolve, it can be coupled with systems dynamics modelling where the
dynamics of a larger system is analysed (Tako and Robinson, 2012). Competence cell networks is
a model in which interfaces and collaborations are modelled between for example different
competence cells (Müller, 2006).
All in all, we conclude that production planning and control research is suitable, as can be seen in
chapter 5, to help an ambulatory care service provider in Sweden overcome its most pressing
operational challenges: achieving a high degree of flexibility in operations as well as
understanding and modelling the environment the unit is acting in. The research trends we have
observed and described above thus make production planning and control theories more relevant
to an ambulatory care service provider, since these aim to tackle the same challenges that
53
ambulatory care service providers are facing. Indeed, it seems like management theories within
healthcare are often built on theories and concepts that were applied first in other industries; where
the Toyota Production Systems (TPS) is a prominent example of this (Sugimori et al., 1977). An
application and utilisation of production planning and control theories focusing on systems
dynamics, networks and holistic approaches towards production planning and control in a
healthcare setting would therefore be in line with the previous academic progression.
6.1.2 ACTIVITIES OF PRODUCTION PLANNING AND CONTROL IN AN AMBULATORY CARE
SERVICE PROVIDER
Since ambulatory care service providers typically are part of larger processes in a hospital in
general (Hulshof et al., 2012), they typically need to handle both referrals and patients in similar
ways from an operational perspective. The process in a radiology department is therefore likely
similar in a number of ways to that of other ambulatory care service providers. This section
analyses and discusses the activities of production planning and control in a typical ambulatory
care service provider in an emergency hospital in Sweden.
The understanding of the process in the radiology department has been central to analyse the
potential for production planning and control in the setting. The process has been analysed and
mapped in the context of the most typically discussed production planning and control activities
discussed in theoretical research (Buzacott et al., 2012; Jonsson and Mattsson, 2003; Slack et al.,
2010). The analysis has been performed to investigate the feasibility of utilising the mindset and
tools related to production planning and control. An analysis of the constraints that limit all
production planning and control activities is therefore also elaborated upon.
6.1.2.1 Loading
Loading is an activity of production planning and control that deals with how much to do in a
workplace and e.g. how much should be done by a specific machine. Loading can here be mapped
by the steps “incoming referral”, “prioritisation” and “scheduling” in the generic process outlined
in chapter 5. The tools the department has at its disposal to affect loading for the department as a
whole are: Deterring the writing of referrals with increased prices for examinations, or to filter
incoming referrals and send these to other radiology departments in the county. Filtering of
referrals is however only done during on call duty hours. The practical applicability of managing
the demand through raising prices for examinations is only applicable to a certain extent and only
to patients arriving from inside the hospital and furthermore likely infeasible in the long run.
It is however possible to affect and level the loading between different modalities, in different
steps in the process: In the prioritisation, the physicians (and in some cases nurses) may choose to
both manage the load on different modalities by referring patients to other modalities. As the goal
is to give the best care available though, it may not always be possible to switch modalities for an
examination.
During scheduling of examinations it is also possible to affect the loading for later stages in the
process, but mostly by offloading different modalities by scheduling the examinations elsewhere.
The step in the process where summons are sent out to patients is another opportunity to let patients
prepare themselves and not the department, thus reducing the load on employees. This is already
utilised and does not change the loading from a volume perspective but frees more time for other
activities.
The loading for the department as a whole can thus only realistically be managed by referring non-
emergency patients to other radiology departments and increasing the waiting times for scheduled
54
examinations. External factors such as budgeting in the hospital and public policies therefore
dictate, top down, the amount of resources and the capacity demands put on the radiology
department (Van Den Berg, 1999). The effects are therefore only implicit and does not affect
loading at its source.
The situation is typical to situations for ambulatory care services: The demand and budget is
dictated top down, and budgetary constraints dictate available resources to an even greater extent
than demand structure. An ambulatory care service provider in a major emergency hospital has to
deal with infinite loading and therefore need to cope with demand rather than plan how much to
do.
Lean and JIT concepts could be seen as solutions since the concepts are, among other things, means
to manage demand and handle situations where different steps of the manufacturing process have
little possibility to affect and forecast their work load. As stressed in literature, these does however
require a holistic approach and not only application in parts of the process (Radnor et al., 2012).
The belief that an organisation should focus on adapting to its environment and not focus on
attempting to affect demand and external pressure has, as mentioned above, been widely accepted
and is an example of a systems dynamics and discrete event simulation approach (Tako and
Robinson, 2012). These studies advocate understanding and a data driven approach towards
understanding the environment and planning according to different contingencies. For this reason,
a further emphasis needs to be put on flexibility of operations (Vlachos et al., 2007). This seems
like a reasonable approach and internal loading and planning should aim to handle all eventualities.
All this however, in turn, requires an approach and a mindset in the entire operations and not only
in loading activities.
6.1.2.2 Sequencing
Sequencing involves deciding the order in which production should be conducted. Prioritisation
of patients in an emergency hospital is indeed the text book example of sequencing (Slack et al.,
2010). In the interviews with physicians, this was identified as the point where physicians are able
to affect planning of examinations.
This prioritisation is done with the sole emphasis on prioritising in accordance with medical
urgency in order to enable the treatment of the most urgent patients first. As such prioritisation is
infeasible to optimise the production planning and control of operations, which is unfortunate since
it would be a great opportunity. The prioritisation could be done in a way that would for example
help the scheduling of examinations of the same organs after one another to reduce set up times
between examinations. This is an example of a major paradigmatic discrepancy in management
studies of healthcare; between management theories and medical professionals (Andersen et al.,
2014; Hans et al., 2012; Zijm, 2000). The situation is especially problematic from a managerial
perspective in the context of healthcare, and ambulatory care services in particular, since the
operations are driven by specialists (Clegg et al., 2013; Hans et al., 2012; Vissers et al., 2001).
This can also be said to be an example of one of the difficulties of planning healthcare activities,
namely that the patient is both involved in the process while at the same time being the customer
(Hall et al., 2006).
Achieving and maintaining an uninterrupted flow of patients and referrals in a sequencing stage is
typically a primary concern in academic studies of radiology department (Kruskal et al., 2012).
Theories suggest different approaches towards achieving this. In Lean a holistic approach with
clear communication as well as knowledge of bottlenecks and wasteful activities among staff and
55
management is used (Womack and Jones, 2010). In the empirical example at the radiology
department at Danderyds sjukhus however, it is contingent on the individual scheduling. Some
have process flow in mind in the scheduling whereas others focus solely on medical urgency.
The situation in the radiology department at Danderyds sjukhus, where the prioritisation is handled
by physicians based purely on medical reasoning is troublesome from a production planning and
control perspective; a lot of the control is relinquished and the operations strategy is at risk of
being neglected. A vital first step would therefore be to achieve knowledge and understanding of
existing problems widespread in the department, in accordance with the Lean reasoning.
6.1.2.3 Scheduling
Scheduling is the planning of when to perform certain activities. The scheduling of appointments
can indeed be said to cover the planning needs in the general process at a radiology department.
That activity does however only pertain scheduling of when to perform examinations short or
medium term, and not long term scheduling of resource utilisation. Scheduling however also
involves staffing and is made more difficult under uncertain demand. The healthcare setting is
therefore especially hard to staff from a production planning and control perspective, since the
competence requirements are diffuse and hard to anticipate.
The capacity of the radiology department is constrained primarily by two resources: number of
machines available and availability of competent staff. This goes well in hand with theory on
healthcare where resources are considered to be the most important limitation in an healthcare
operation (Blake and Carter, 2002; Vos et al., 2007). Because of this the scheduling of production
planning and control can be said to mostly concern the scheduling of non-emergency patients. This
study has, as mentioned in the delimitations section of the introduction, disregarded long-term
capacity planning.
Understanding of historical demand is an important part of the planning process (Clegg et al.,
2013; Melnyk et al., 2014). The empirical study in this thesis would however suggest that this
understanding is based on indicative cumulative experience of employees in the radiology
department at Danderyds sjukhus. The situation was similar, prior to the production planning and
control initiative, in the benchmarked radiology department at “Universitetssjukhuset i Linköping”
which is a situation that was echoed by the benchmark. We however believe that the department
could be better equipped to handle external events and changes in demand by extending the
understanding of demand data. This would enable the adaptation of scheduling of operations to
anticipated changes in demand without requiring the accumulation of experiences, which supports
this view (Kopach-Konrad et al., 2007).
Scheduling can also be seen from the experiences after the implementation of a new production
planning system in Linköping; the planning is condensed and highly data driven using historical
demand. Patients not required to come in for emergency appointments are scheduled as soon as
possible, and not prioritised further in another step as in the radiology department at Danderyds
sjukhus. At first the staff was opposed to the change, which is in line with theory (Andersen et al.,
2014; Kopach-Konrad et al., 2007). After being convinced following the first results obtained from
using the new data driven approach both the staff morale improved as well as the throughput from
the department.
6.1.2.4 Monitoring and Control
Monitoring and control activities performed in the department are done as standard parts of the
generic process. Primary reads and diagnoses are always controlled and verified by double reading
56
physicians and rounds. These activities aim to avoid clinical mistakes and to make sure that the
production; diagnosing and treatment of patients, involves no mistakes; avoiding disruptive
mishaps that affect the operations are thus not prioritised. The empirical study would however
suggest that disruptive operational mistakes may in affect the clinical treatment of patients since
resources are utilised in vain.
One primary goal of performance measurements is to aid the monitoring and control of the process
on a continuous basis. Monitoring and control from a production planning and control perspective
is thus made harder due to the lack of measurement systems both on a national and department
level (Melnyk et al., 2014). The goals stipulated in the radiology department at Danderyds sjukhus
are furthermore constructed for annual feedback. This disables live monitoring and furthermore
disallows ad hoc resource allocation top down.
Recent theories on performance measurements advocate the usage of continuous feedback loops
and a data driven approach with the spreading of understanding among staff. Theory also point to
the need of monitoring when implementing change to an operation which further highlights the
importance of improvement (Melnyk et al., 2014; Micheli and Mari, 2014). This would suggest
that a clear area of improvement for the department, as echoed above a, would be a data driven
approach with relevant short term measures.
We therefore think that an efficient monitoring and control of the operations in an ambulatory care
service provider can be achieved through an increased and recurring usage of relevant performance
measurements. One should remember that it is not just something that is implemented, it is vital
to define targets representative of the operation (Bhasin, 2008). Stipulated performance
measurements should thus aim to mitigate both medical and operational objectives, as mentioned
above in the discussion on loading. This goes in line with the improvement in monitoring and
control that was a major part of the new method implemented in Linköping.
6.1.3 SUMMARY AND CONCLUSIONS
The analysis of the activities of production planning and control described by academia illustrates
that a typical process in an ambulatory care service lends itself well to being utilised in the generic
process flow in the radiology department at Danderyds sjukhus. Our analysis also shows that
production planning and control theories aim to solve and handle problems that are common in a
typical ambulatory care service provider. We therefore think that prevalent contemporary research
in production planning and control is highly applicable in the setting of an ambulatory care service
provider. However, successful implementation is highly contingent on creating awareness and
promoting an environment where experience is encouraged and shared. A systematic approach
incorporating both a systems and network perspective, based on historical data, would be a
reasonable initiation.
We conclude that the process in the radiology department at Danderyds sjukhus currently
encompass production planning and control activities. However, the activities performed are done
so without explicit regard to the overall operations strategy. For this reason, implementing
production planning and control is feasible in the daily operations. An implementation of
production planning and control methodology and reasoning however require an additional regard
and mindset in the regular operations. However, the case studies conducted in this thesis also
illustrate that a production planning and control mindset needs to be preceded by a thorough
mapping of the regular processes. Otherwise it would be hard to identify and map stages or steps
in the process that would lend themselves to production planning and control activities.
57
We conclude that the overall goal of operations in a radiology department in a major emergency
hospital in Sweden is first and foremost described as examining and diagnosing patients
qualitatively and to meet the needs of all individual patients. Performance measurements for the
radiology department as a whole do however not encompass only these objectives explicitly.
Instead, the measurements are implicit; derived from other departments’ goals, and at times
counteractive. The connection between operational metrics and clinical activities furthermore
needs to be improved. The situation with implicit measurement systems that does not fully
represent the objectives of the organisation is a situation that is likely typical for ambulatory care
service providers in a more general sense. In these cases production planning and control need to
function as reconciliation between counteracting goals.
The clinical objectives of the department obviously dictate the operations strategy, since that is
their raison d’être. These medical objectives affect the work in different ways but ultimately set
up boundaries for operations. Special emphasis is therefore to reduce the waiting time for
emergency patients because of medical severity and the risk for increased morbidity. This would
not have been a concern in traditional manufacturing industries, where these activities simply
could have been disregarded.
Department goals can however be interpreted and adapted to a production planning and control
context, which would suggest that implementation is feasible in the current operations. We have
concluded that the objective of a radiology department in operations management are:
Maintaining a smooth flow.
Working efficiently and eliminating waste and unnecessary activities.
Achieving a good working environment partly by eliminating stress.
These need to be reconciled with budgetary and demand constraints as well. The radiology
department’s operational constraints furthermore mainly relate to machine capacity and access to
competent personnel in line with general resource constraints in a healthcare operation (Blake and
Carter, 2002; Vos et al., 2007). Compared to other hospital departments the radiology department
is clearly distinguished by the safety issues concerning radiation levels.
6.2 PROCESS FLOW IN RADIOLOGY DEPARTMENT
Process flow has been identified as a main concern in the regular operations in the department.
Maintaining a smooth flow is imperative for a functioning department that needs to handle flows
from many different instances, with different contingencies and needs, on a daily basis (Kruskal
et al., 2012).
A source for complexity in the process of conducting examinations is the fact that a radiology
department; an ambulatory care service provider, in a major emergency hospital in Sweden need
to handle different types of flows in the daily operations. An example of this is that all
examinations require the department to handle both referrals and patients in two separate flows
that need to be coordinated. The flow is furthermore highly contingent on the medical urgency and
thus priority of the patient. Emergency patients and non-emergency patients can be considered to
be two inherently different flows. It should be noted that this situation is applicable to any
ambulatory care service, that both needs to handle and administer referrals and also both
emergency and non-emergency examinations.
6.2.1 REFERRALS AND PATIENTS FLOWS
58
Referrals are central in radiological practice and concerns two way communication and
information exchange between referring physicians and the radiology department. Our empirical
study suggests that referral communication is mostly handled in internal IT-systems in healthcare
nowadays. Patient handling concerns all handling of different patients both in correspondence and
direct communication, as well as in examinations and treatments.
The radiology department need to treat these flows differently due to their inherently different
nature and issues related to them. The referral and the patient flows require coordination since they
are contingent on one another in different stages.
6.2.1.1 Referral Flow
The referrals aim to facilitate a JIT-system where care is given when needed and a flow of relevant
information on a need to know basis in the system. For this reason they can be said to be essentially
equivalent to Kanban cards utilised in Lean and JIT systems (Persona et al., 2008). Our empirical
study shows that referrals handling entails different problems that are mostly connected to
insufficient information being forwarded which in effect leads to unnecessary lead times and
activities. The magnitude of the effects are however dependent on the competence and experience
of radiology nurses and referring physicians.
Insufficient referrals may result in unnecessary stoppages and activities, which inhibit both the
patient and referral flow, as exemplified in chapter 5 of this thesis. They may also lead to avoidable
radiation exposure for patients due to unnecessary examinations, which is a safety issue.
Insufficient referrals are examples of sources of production inefficiency and give rise to
inefficiencies in the process and do not only affect the flow in the radiology department, but also
for all departments referring patients to the department. This illustrates a typical situation that the
Lean philosophies aim to improve on; it is vital to uncover and identify these actual problems in
order to be able to handle the symptoms they create in other parts of the organisation (Ahlstrom,
2004; Andersen et al., 2014; Womack and Jones, 2010).
We believe that the issues relating to mistakes in the referrals handling are symptoms of poor
understanding in the referring departments of the radiological practice and a lack of a sense that
the different departments of the hospital are part of the same organisation as one another. The
decision to hire and educate specialised emergency physicians at Danderyds sjukhus is promising
from a production planning and control perspective, since it will enable the preservation of
competences and experience in referring to the radiology department. It will also likely enable
closer collaboration between the radiology department and the emergency care unit. This has the
potential to have widespread positive effects given the large fraction being referred for emergency
examinations to the radiology department.
In the case study, the managerial emphasis in the hospital has been put on improving competence
in the emergency department and to write quick reference guides. Impacting ways of working,
especially long term, in other departments has however proven difficult, much due to employee
turnover. As identified earlier, this is a specialist environment averse to changes, especially ones
from outside actors (Andersen et al., 2014; De Vries et al., 1999). It is therefore not certain that
quick reference guides will be read by referring physicians.
The referral handling in the radiology department at Danderyds sjukhus can be described and
understood from different theoretical perspectives. On the one hand this can be understood as a
situation where Kanban cards are misused. It can be seen as a system whereby the referrals (or
Kanban cards) on cyclical demand environment, as in the case of the majority of the demand in
59
the radiology department, create delays inherent in the structure of the system (Ebrahimpour and
odarress Fathi, 1985). There are theories on how to improve communication and evolve
information flow in the system; Sugimori et al. propose that Kanban cards have the potential for
increased efficiency if used correctly; Joosten et al. as well as Persona et al. Kanban cards are
means to smoothen flow (Joosten et al., 2009; Persona et al., 2008; Sugimori et al., 1977).
This can also, from a Lean perspective, be seen as a situation where Lean tools are applied, but
not utilised system wide which results in failure to align with overall strategy (Radnor et al., 2012).
The situation can likely be improved by a holistic approach and the inclusion of referrals handling
in the overall operations strategy of the operations. Measures to improve effectiveness, or the
practice of doing the right things as an organisation, have furthermore been studied by many
researchers. The prevalent notion to approach production planning and control of referrals from a
holistic perspective is therefore likely applicable (Nguyen et al., 2014).
To conclude, we conclude that emphasis should be put on increasing the understanding of
radiological practice in general and the radiology department in particular in referring departments
in the hospital. This goes in line with studies of the healthcare setting in general that advocate team
member development to, for example, reduce the risk of failures (Kilgore and Langford, 2009).
The referrals are not affected by any external forces and can therefore be managed by the
department. We therefore think that all unnecessary activities in the hospital stemming from
inefficient referral handling are avoidable. Insufficient referrals will, as described above, result in
the need for unnecessary activities in the patient flow, and the two flows will therefore need to be
handled with outmost care. Indeed, there is great potential to improve operational effectiveness
and efficiency by assuming a more systematic approach towards managing referrals.
We also think that this solution can be extended to any ambulatory care service provider in a
hospital; any ambulatory care service provider will benefit from promoting understanding and
competence in its referring departments. This is illustrated both in and outside of the healthcare
setting where Kanban and JIT techniques improve coordination and communication
(Ciemnoczolowski and Bozer, 2013; Köchel and Nieländer, 2002; Matzka et al., 2012; Olaitan et
al., 2014; Persona et al., 2008). As will be discussed later the collaboration between departments
may vary however which may both hamper and promote operational excellence.
6.2.1.2 Patient Flow
Patient handling constitutes a relatively limited part of the process in a radiology department. In
order to facilitate a good flow through an ambulatory care service provider it is however imperative
that the examinations are done efficiently and effective. In order to achieve this the whole process
of handling both patients and referrals need to be executed appropriately.
In the empirical study in the radiology department at Danderyds sjukhus it became clear that some
issues relating to production planning and control were specific to different modalities whereas
some were prevalent in the generic radiology process. An example of this is that CT may lag due
to new personnel not accustomed to the tempo and mindset. CR experiences interruptions in the
flow disproportionally often because the images produced are insufficient for diagnosis.
Furthermore, technology improvements in different modalities, such as the possibilities for
ablation procedures, are not accounted for in the planning process. This highlights the fact that
there are different planning methods needed for different modalities, especially since they become
more and more specialised. We also think that the challenge of managing systematisation on a
department level with increasingly specialised subunits is a typical situation in healthcare systems
today. From a production planning and control perspective, this situation needs to be understood
60
thoroughly both in terms of what can be done (Müller, 2006) and how situations may develop
(Tako and Robinson, 2012).
Satisfying individual patients’ needs and preferences in a generic process can be said to be a major
challenge for all healthcare systems; a wide range of academic studies describe patients as one of
the most influential environmental variables affecting operations (De Vries et al., 1999; Kruskal
et al., 2012). Since the patient is both the customer and, in a sense, the product, the process needs
to be highly contingent on the characteristics of the patient (Nguyen et al., 2014). There are several
theories suggesting different approaches on how to handle patients most efficiently. Some suggest
building a robust operations strategy that handles all patient events, and using a systems dynamics
or a discrete event simulation approach depending on the timeframe to model eventualities (Tako
and Robinson, 2012) and thus perceiving the healthcare organisation as a network of competence
cells and base the adaptation of tools, methods and principles of production planning and control
accordingly (Müller, 2006). This is important to keep in mind when applying any management
theories in a healthcare setting.
We have identified several recurring issues that can be seen as avoidable. Many of these stem from
interruptions in the flow of patients and have both clinical and operational implications. These
interruptions in the patient flow are symptoms of a number of internal and external issues relating
to individual errors and ultimately competence inadequacies. We furthermore think that the
theories of Müller regarding the healthcare organisation sorted by competences collaborating in
non-hierarchical ways are applicable, and that many issues are likely to improve with the systemic
change in the emergency department by hiring specialist emergency physicians (Müller, 2006).
6.2.2 EMERGENCY AND NON-EMERGENCY FLOW
The radiology department categorises patients and prioritise examinations of patients according to
medical urgency. In a broad sense patients can be said to be either emergency patients or non-
emergency patients. Emergency patients require immediate, or almost immediate, examinations.
Non-emergency patients can on the other hand be scheduled in the future, sometimes with no other
restrictions than the national guidelines for waiting times for different examinations. Non-
emergency examinations are thus more easily planned, since they offer temporal flexibility to the
department.
The flow of emergency patients directly affects the capacity to schedule non-emergency patients,
and in turn the waiting times for the latter. Initiatives to disjoin the two flows, by for example
specialising machines on either emergency or non-emergency examinations, have proven effective
in both Danderyds sjukhus and in “Universitetssjukhuset i Linköping”.
Studies show that it is hard to operate using Kanban cards in a multiple product environment and
require special attention and coping strategies (Olaitan et al., 2014), which is highly applicable in
this setting. We consider the demand profiles to be highly contingent on whether it is emergency
or non-emergency, where the two examination types need to be treated differently. One typical
distinction between the two flows is due to the fact that the two flows are categorised depending
partly on the type of examinations and the state of the patients; more critical diagnostic queries on
patients with more serious medical state are typically classified as emergency examinations. The
emergency flow is also inherently uncertain and more or less needs to be handled instantly by the
healthcare provider. This, in combination with the individual needs of patients (as discussed
above) inhibits a comprehensive anticipation of all activities that need to be performed. Our
empirical studies on the other hand show that non-emergency examinations are far less exposed to
uncertainty and unanticipated events. Olaitan et al. also suggest that the natural way to handle
61
multiple products is to treat different products as separate manufacturing lines (Olaitan et al.,
2014). The notion to separate operational processes based on their differences is also echoed in
other operations management studies, such as in enterprise resource planning by van Merode et al.
(van Merode et al., 2004). The emergency flow will always be present in an emergency hospital,
and the non-emergency flow will therefore have to be organised accordingly. Scheduling offers
the possibility to plan the production more carefully but also takes time and resources.
In order to adapt, the non-emergency academic research would suggest a discrete event simulation
approach to handle short-term changes in the emergency flow (Tako and Robinson, 2012). This
illustrates the further need to fully understand the situation and the need for continuous feedback
in an ambulatory service provider in particular.
6.2.3 SUMMARY AND CONCLUSIONS
An ambulatory care service provider in a major emergency hospital in Sweden acts in an
environment where it is infeasible, impossible even, to eliminate any of the flows it is hard to deal
with the situation in the same ways as in traditional manufacturing industries. We however think
that an ambulatory care service provider should focus on separating the flows as much as possible
and let the different flows act in an environment where the conflicts and issues do not arise. An
example of the separating approach would be to specialise machines to handle specific flows:
emergency or non-emergency patients. This goes in line with other studies conducted in the
hospital setting; e.g.(van Merode et al., 2004).
Ambulatory care service provider should act on promoting closer collaborations and interaction
between departments, and in this case to improve referrals handling. The referral flow is clearly
the flow with the least exposure towards hospital external factors and contingencies. We therefore
think that the aim should be to solve as many issues in this flow as possible on a hospital level. An
improved coordination in the referral flow furthermore opens up the potential for improved
coordination between departments in the hospital. Improved coordination is always beneficial to
an ambulatory care service provider since the unit is just a part of the patient’s journey through the
hospital. In a “Lean organisation” management should aim to identify and uncover waste in the
process as much as possible, since it will then be possible to deal with the problems head on
(Sugimori et al., 1977; Womack and Jones, 2010). We therefore think that since Danderyds
sjukhus strive to be a Lean organisation improved referrals handling should be an area of focus for
the hospital as a whole.
Based on our empirical findings and the opinions of the interviewees we furthermore think that
there is a clear case for separating the handling of emergency and non-emergency examinations.
The non-emergency flow is subject to limited uncertainty, but the potential to utilise this in
production is inhibited by the involvement of the highly variable emergency flow. The separation
of the emergency and non-emergency flow has proven useful in the radiology department at
“Universitetssjukhuset i Linköping”, as well as for a limited time in the CT modality in the
radiology department at Danderyds sjukhus.
We furthermore think that the department should focus on mapping the activities that are recurring,
but cannot be anticipated, so as to not be too dependent on the experiences of individuals. We also
think that there is potential in letting patients choose times for examinations, letting nurses perform
62
some tasks currently performed by physicians8, and to plan examinations more explicitly in the
prioritisation step.
6.3 COLLABORATION AND COORDINATION
Hospital integration that enables smooth flows and interactions between different units is a widely
discussed topic (Hans et al., 2012; Harper, 2002). The contention to treat hospitals as single
entities, and not as a group of individual organisations acting in a large market system, in order to
avoid situations where departments see themselves as separate systems with other objectives than
the whole hospital, is prevalent in academic research (Hans et al., 2012; Roth and Dierdonck, 1995;
Vissers et al., 2001). This entails collaboration between both different work groups with, and
different departments in the hospital, as well as coordination of medical equipment and hospital
facilities (Blake and Carter, 2002; Vos et al., 2007).
We have seen that the topic of collaboration and coordination of resources both within the
department and in interactions with other departments has continuously been a subject for
discussion both in public opinion in Swedish healthcare and in theoretical research (e.g. (Kilgore
and Langford, 2009)).
6.3.1 INTERNAL COORDINATION
The resources available in public healthcare in Sweden are mostly determined by public policy
both on nationwide governmental level and in the local County councils. Due primarily to this the
main objective for an ambulatory care service provider in a major emergency hospital in Sweden
cannot be revenue growth. The focus of coordination should therefore be optimise the utilisation
of resources currently available and not assume the potential for acquisitions (Blake and Carter,
2002).
In the empirical study, the main area of internal coordination of resources takes place in the staffing
of nurses and physicians. Collaboration between different modalities also aims to improve the
overall operations of the department through collaborative sharing of operational learning. The
typical healthcare environment contingent on the coordination of different specialists is another
widely discussed topic in academic research (Clegg et al., 2013; Kopach-Konrad et al., 2007;
Radnor et al., 2012)
6.3.1.1 Staffing
One of the main determinants of operational capacity in healthcare is the availability of scarce
resources in terms of staff, and the availability of medical equipment (Blake and Carter, 2002;
Vos et al., 2007). We have found this to be true for a radiology department in a major emergency
hospital in Sweden. Indeed, staffing should be considered as the main area of coordination and
venue for interaction between different operational units inside an ambulatory department. The
availability of competent personnel is, together with the machine capacity, the main determinants
of production capacity in a radiology department. Staffing should therefore aim to optimise
operations according to the former constraint.
Rotation of personnel stems from the demand of covering the competence requirements in
different modalities at all times, and the common problem of extreme variation in daily workload
(Beliën and Demeulemeester, 2008). An absolute number of individuals need to have basic
proficiency to operate the CR and CT machines during non-office hours, which needs to be
achieved through hands on practice in the different modalities. The large competence
8 Such as prioritisation of some examinations
63
dependencies in different modalities furthermore underline the need for spreading of knowledge
in the department through rotation. Some radiology nurses furthermore feel that the work tends to
become tedious without variation from rotating between different modalities.
Rotating between different modalities may however inhibit the development of specialist
competences. The rotation may also lead safety issues if physicians or nurses lack competences.
By not rotating, one however creates situations with competence dependencies in the different
modalities which, in turn, creates a situation where the process is driven in large by individual
members of staff (De Vries et al., 1999).
The fact that the staffing of radiology nurses and physicians are handled separately in the radiology
department at Danderyds sjukhus and that certain competences are locked to persons and not role
creates staffing dilemmas. Rotating schedules make coordination even more difficult and
relinquishes further control over managing the department from management to individuals. This
creates further issues especially in modalities where both professions are needed when conducting
examinations.
In the radiology department at “Universitetssjukhuset i Linköping” the department organises
staffing in different groups of radiology nurses that are not tied to one specific modality. These
teams are supposed to be able to operate and execute the most common examinations in all
modalities, but also to be able to execute more complex procedures on specific organs. In
production planning and control terminology these sections would be labelled multifunctional
teams. This decentralisation of responsibilities is often adopted in Lean initiatives on a workforce
level (Ahlstrom, 2004). We think that this is a reasonable approach, both to achieve operational
collaboration in the department, and also to maintain a staff group less exposed to employee
turnover.
We think that rotating schedules are necessary to facilitate efficient planning in any healthcare unit
with different subunits, since it is the only way to ensure the hands on experience required to fulfil
competence requirements. Competence dependencies need to be remedied by staffing. Production
planning and control becomes more difficult with separate scheduling of physicians and nurses,
and this need to be developed so as to create a scheduling process that does not result in preventable
issues. Our empirical study however illustrate that some rotation might be unavoidable since the
flow of patients to for example the modalities fluoroscopy and angiography is too small to allow
for developing the competence for all staff in the department.
Even if the nature of the work for radiology nurses and physicians is inherently different they can
both be seen as specialised competences that have a lot of interaction and collaboration. For this
reason we think that it is inadvisable to separate the two in for example a competence cell
networks, as outlined by Müller (Müller, 2006). The coordination of schedules does in no way
mean that physicians and nurses should have exactly the same schedules, but rather that the
scheduling should be coordinated systematically to ensure sufficient competence staffed at all
times.
6.3.1.2 Internal Operational Collaboration
Currently, the only non-clinical collaboration and coordination between modalities in the
radiology department at Danderyds sjukhus is, apart from the shared staffing base, mostly
conducted through meetings between modality responsible nurses on a discontinuous basis. The
department has, as mentioned earlier, not identified any operational needs for continuous and ad-
hoc collaboration on a day-to-day basis between the different modalities. It is hard to coordinate
64
and adjust the quantity of staffing even though a common problem in hospitals is the extreme
variation in daily workload for nurses (Beliën and Demeulemeester, 2008). On the contrary, the
department have concluded that clinical activities might be inhibited due to too many people being
involved in conducting single examinations.
As mentioned earlier it is central in a Lean organisation to identify problems and promote
decentralised responsibilities. A way to achieve this is by informal and frequent short meetings,
where learnings and issues can be shared and discussed among staff. These are called “daily stand
up meetings”, as used in the Toyota Production Systems approach. This promotes not only
communication, but also helps the organisation to promote feedback and individual accountability
(Middleton and Joyce, 2012).
We think that the operational collaboration and a continuous collaborative operational learning
between modalities need to be extended and more systematic in order to facilitate further
improvements. The approach with instituting stand-up meetings weekly or monthly on different
levels: between radiology nurses responsible for different modalities, in different modalities, and
with physicians and radiology nurses, sounds reasonable and applicable in this case. Promoting
internal collaboration in such a way is likely applicable in ambulatory care services in general.
6.3.2 INTERDEPARTMENTAL COORDINATION
An ambulatory care service provider does not follow patients for their entire treatment and
diagnostics in the hospital. Efficient coordination is therefore central in the collaboration around
logistics planning and division of responsibilities, as noted repeatedly. We therefore see inter-
department coordination and collaboration as a major factor in the success of an ambulatory care
service provider in a major emergency hospital in Sweden.
Well-functioning logistics is an area that is central in the coordination between departments and
are imperative in achieving and maintaining a smooth flow through an ambulatory care service
provider (Kruskal et al., 2012). Patient transports, and the coordination between two different
departments, as well as the janitorial service are indeed messy and dependent on experienced
nurses at this point. We believe that the technological solution of handling transportation via
telephone calls and not via central IT-system leaves the potential for improved efficiency in the
janitorial department. Another recurring issue is that patients are not always prepared for
examinations upon arrival, which creates avoidable stoppage times in the labs. This can be likened
with the theoretical development of production planning and control where IT seems to have
played an increasingly central role in different methods and strategies (Olhager, 2013). Our study
suggests that interactions and coordination between departments differ between modalities and
can be said to be rather dependent on individuals. Overall we have observed that coordination
between departments work best between departments were the employees have managed to
establish personal connections. In our empirical study many interviewees noted, as seen above,
that referring departments with contact persons that coordinate interaction and referrals were easier
to coordinate with. As discussed above, the empirical study (emergency ledger) also suggest that
formalised and standardised communication and coordination by utilising the full potential of the
IT system is preferable.
The case study furthermore shows that division of responsibilities between departments is another
area requiring interdepartmental coordination that becomes especially apparent when it comes to
patient logistics. Decisions on who is responsible for the safety or safekeeping of the patient need
to be made on an inter-department level and should account for example different possible medical
emergencies as well.
65
Patient treatment in emergency hospitals can from a theoretical perspective be seen as large JIT
systems coordinated with Kanban cards. The Kanban cards in this case are, as mentioned above,
referrals (Ahlstrom, 2004). The different units and their interaction in the hospital can furthermore
be viewed from a systems dynamics perspective or focus more on the interactions with a more
network oriented approach (Müller, 2006; Tako and Robinson, 2012). Ciemnoczolowski and
Bozer conclude that the handling of Kanban cards and the size of the batches “have a bigger impact
on workstation starvation than the prescribed cycle time” (Ciemnoczolowski and Bozer, 2013).
Since an ambulatory care service unit does not have its own end product and are rather parts of
other systems, it is particularly important to see how it can utilise Kanban cards to smooth flow
and avoid waiting times for other stations in the process.
We think that the emphasis on understanding and modelling events rather than forecasting demand
will likely make communication and coordination of productivity goals become more important
in today’s healthcare systems. The uncertain demand, in combination with the individual
contingencies of individual patients means that this view is especially applicable in a healthcare
setting. Understanding the structure of organisations and the division of responsibilities is indeed
a major challenge in the 21st century, but does require a deeper understanding concerning which
operations management tools that should be applied (Müller, 2006).
The radiology department is developing by designing a department that is partly allocated within
the emergency care unit, from a production planning and control perspective (Olaitan et al., 2014).
This will have several benefits from a production planning and control perspective but, most
importantly, it will promote collaboration and competence exchange between the departments as
well as simplify logistics. There is indeed room for waste minimisation and facilitation of
discussions to prevent unnecessary activities in both ends of the referral flow.
6.3.3 SUMMARY AND CONCLUSIONS
We conclude that coordination within an ambulatory care service provider with different processes
is central to efficient operations. We think that there is a potential in coordination of resources to
reduce waiting times on a department level rather than for different modalities, as shown in the
benchmarking study. However, in order to fulfil the demands that is put specifically on a
department in an emergency hospital interdepartmental coordination and collaboration is key.
Currently, interdepartmental coordination and collaboration typically encompass medical
diagnosing and treatment, but should also handle operational issues such as logistics of patients.
Many of the problems identified in the referral flow can be said to be connected to poor
collaboration within the hospital. We think that strategic production planning and control to
coordinate and collaborate both within the department and between departments in the hospital
need to be extended and reviewed continuously; it will otherwise be hard to handle new operational
challenges in the future. Collaboration indeed needs to be reviewed continuously – personal
connections are dependent on continuous contacts. The continuous development of all personnel
is also central to goal fulfilment and risk reduction (Kilgore and Langford, 2009).
We furthermore think that there is a lot to utilise, on a hospital management level, from theoretical
reasoning and modelling in production planning and control. It is vital to model and understand
interactions and dynamics between different units in a hospital in order to be able to perform
operational changes hospital wide.
66
7 CONCLUSIONS This chapter will present the final conclusions of the work, how the different research questions
presented have been answered and ideas of future topics to study.
This thesis has had both theoretical and empirical purposes. This chapter will focus primarily on
the theoretical purpose of generalising the case study findings from a radiology department in a
major emergency hospital in Sweden in a more general production planning and control
perspective for an ambulatory care service provider. The empirical purpose of this thesis pertaining
to the studied department will be discussed in “Section 7.2.1 Recommendations to the Radiology
Department at Danderyds Sjukhus” more specifically.
In order to achieve the theoretical purpose, a main research question was formulated: “How can
production planning and control aid and improve the work process of an ambulatory care service
provider in a major emergency hospital in Sweden?” and three sub-questions posed to help answer
the main research question:
1. How can the work process in an ambulatory care service provider in a major Swedish
emergency hospital be mapped out from a production planning and control perspective?
2. What general challenges and opportunities exist for an ambulatory care service provider
in a major emergency hospital in Sweden from a production planning and control
perspective?
3. What activities should an ambulatory care service provider in a major emergency hospital
in Sweden prioritise in order to improve the situation using techniques, methods and
principles of production planning and control?
7.1 ANSWERS TO RESEARCH QUESTIONS
How can the work process in an ambulatory care service provider in a major Swedish emergency
hospital be mapped out from a production planning and control perspective?
The mapping performed in this study is based on the collective knowledge of staff in the radiology
department at Danderyds sjukhus. The documentation on the production process, outlined in
Chapter 5, was however somewhat lacking, which could indicate that process orientation has not
been a major focus previously. From our literature study, and the empirical investigation, it is
suggested that this is a common problem in the context of healthcare. It is thus likely applicable to
ambulatory care services in general.
The process flow is important to map in order to get an overview and a baseline for further
implementation. The process in the radiology department, as mapped in Chapter 5, is indeed
unique from a production planning and control perspective depending on how one looks at it. We
found that it could be seen as one process with different flows of referrals and patients as part of a
major healthcare system, as in the case of typical ambulatory care service providers. It is also
important to get an understanding of the contingencies and ways of working in different subunits,
since this will impact how the implementation can be executed. We have found in this case is that
the process does not differ greatly between modalities in the radiology department at Dandeyryds
sjukhus, even though there are different contingencies and constraints apparent.
A successful and useful mapping both requires and facilitates shortening of today’s feedback loops
and follow-up activities. As in the example at Danderyds sjukhus; the process oriented and Lean
initiative that the hospital adheres to requires the identification of areas of improvements.
67
However, a lack of quality measures that connect clinical achievements and productivity KPIs
specifically aimed at radiology is a hindrance in this respect.
One can also focus on categorising patients and map the process for either emergency or non-
emergency patients, which would be suitable for ambulatory care service providers on a general
basis. However, from a strictly production planning and control perspective it is most efficient to
distinguish between emergency and non-emergency patients since these flows need to be handled
differently.
Referral flows and interfaces with other departments are also important to manage and understand.
This is especially important for an ambulatory care service unit since it typically handles referrals
in more instances than patients (as illustrated in the process outline above).
The specialities and expertise is important to keep in mind when mapping the process from a
production planning and control perspective. It is possible to get two distinctly different views of
the process by viewing it from a strictly clinical perspective or from an operational standpoint. We
think that a useful mapping of the process requires a holistic view that encompass and takes all
these perspectives into account.
Furthermore, the clinical activities that are complex and difficult to standardise can likely be
anticipated to happen at some point, and there is potential in preparing for different scenarios (Testi
et al., 2007). The utilisation of production data is however important to develop in order to achieve
this (Kopach-Konrad et al., 2007) and would be a useful tool for communicating both externally
and internally, and in turn enable a smooth flow and a more efficient coordination of resources
(Melnyk et al., 2014; Micheli and Mari, 2014).
What general challenges and opportunities exist for an ambulatory care service provider in a
major emergency hospital in Sweden from a production planning and control perspective?
From the mapped out process in Chapter 5, we found that it is indeed difficult to use a standard
framework that fully encompasses the entire process of a radiology department at a major
emergency hospital in Sweden. As can be seen by the different frameworks found in the literature
it is likely a challenge for any hospital, emergency hospitals in particular, to fulfil the underlying
assumptions of the predominant frameworks for production planning and control (Nguyen et al.,
2014).
All activities in the radiological process in the case study department are relatively interconnected
and contingent on one another - and also part of the process. Our empirical studies would suggest
that this is typical to not only radiology departments and ambulatory care service providers but
healthcare units in general. Managing the process and facilitating a smooth flow thus requires the
reconciliation of different activities and a holistic thought process, which in turn will enable a more
efficient fulfilment of objectives (Mazzocato et al., 2010; Nguyen et al., 2014; Radnor et al., 2012).
It is furthermore important to acknowledge that healthcare providers are often governed by not
only operational objectives, but rather need to be adapted to the patient and clinical contingencies.
When trying to improve the production planning and control, a holistic approach is likely required
to reconcile both operational and clinical constraints. The very specific environment does however
not allow for full proof solutions as both demography and external factors influence the situation
greatly (Nguyen et al., 2014; Zijm, 2000).
The process in the radiology department requires documentation; the intersection between
operational and clinical requirements remains unclear and implicit however. This is a challenge
68
from a both production planning and control perspective. Measurements and tools for operational
governance and control are needed to meet and satisfy the increasing demand.
Another challenge facing all of healthcare is that a lot of competence is bound to individuals and
not the organisation. This creates situations where it is harder to manage the department top down,
limiting the power of management. Competence dependencies are indeed prevalent in several
areas that must be dealt with somehow. These are closely related to problems with staffing in
general. For this reason the process is to a great extent driven by medical specialists and the
operations are therefore to a great extent built on experience and culture. We have found, in
accordance with other social studies in the healthcare setting, that the work is built almost solely
on clinical considerations and a reluctance to incorporate management theories.
A diverse patient population that function as materials of production, part of the product, and as a
customer poses clear issue from an operations management perspective. For this reason it is not
possible to anticipate all activities that will be required in the department any given day. However,
many of the idiosyncrasies of individual patients9 can be planned for. It should also be noted that
this does not pertain only to a radiology department, but also to the healthcare setting in general.
It is therefore indeed an opportunity for documenting how to handle and care for different patients
and deliver the most efficient and qualitative care possible.
When it comes to collaboration and coordination of resources we have found that today it is mostly
concerned with matters regarding medical issues. The medical collaboration between physicians
in different departments is extensive and has been for some time. Cooperation in operations
strategies is however not as extensive. From the literature review we conclude that this scenario is
likely apparent in the general healthcare setting as well.
The effects of public policy is rather implicit for ambulatory care service providers. Since
ambulatory care service units are typically not facing the public directly they are not subject for as
much interest in public forums. We believe that the lack of directed quality measures, in Sweden,
specifically for ambulatory care service providers are due to this lack of interest.
The operational resources at hand in an ambulatory care serviced provider are furthermore mostly
dictated by external factors and stakeholders. Indeed, the most important determinants of the
amount of resources available are the availability of competent personnel to hire as well as the
budget to hire competent personnel. A radiology department in a major emergency hospital in
Sweden thus has little power to control the amount of resources to utilise. This is also likely the
case for an ambulatory care service provider in general.
The general challenges and opportunities that exist in an ambulatory care service provider can thus
be said to be rather typical for the general healthcare setting. Typical for ambulatory care service
providers is however the importance of interfaces with other departments, and the implicit effects
from public policies. This presents the opportunity to utilise prevalent management theories that
have proven useful in other settings; with conscious adaptation and understanding.
9 Such as claustrophobic patients, patients with impaired mobility, and patients with mental disabilities etc.
69
What activities should an ambulatory care service provider in a major emergency hospital in
Sweden prioritise in order to improve the situation using techniques, methods and principles of
production planning and control?
The issues above may also be viewed as opportunities. We think that a lot of opportunities lie in a
holistic approach towards production planning and control. Holistic in this respect means to
include both clinical activities and more production related activities in the planning and to achieve
a mindset that permeates the unit. It would indeed be hard to change parts of the process or certain
activities without considering all flows and the overall process. In order to achieve a holistic
approach one needs to create closer ties and a more systematic feedback from key performance
indicators based on both medical and operationally oriented measurements. This requires a
systematic approach with extensive quantitative data gathering and simulations to understand the
demand situation. It seems to us that many changes in healthcare require a sense of urgency in the
organisation in order to create acceptance for change. A change requires different mindsets and an
acceptance for the importance of operational excellence to achieve medical objectives.
The utilisation of production planning theories needs to be based on needs and deciding what is
applicable to the current context is therefore central. Common methods, such as theories
concerning Kanban cards and waste minimisation in Lean operations may not be applicable in all
instances for a wide range of reasons. Indeed, production planning and control is not a toolbox that
you can just put into an organisation and make it work by itself. It needs to be adapted to the
organisation.
An ambulatory care service department could furthermore benefit from focusing on extending the
collaboration both within the department and with other departments in matters concerning
operational issues. It is however important to acknowledge that the challenges that the department
is facing are not symptoms of underlying issues. It is also important to note that the changes
implemented in the radiology department at “Universitetssjukhuset i Linköping” cannot be
mindlessly implemented, a thorough understanding of the applicability needs to be achieved first.
Separating flows and distinguishing between different scenarios is especially applicable in an
ambulatory care service provider. In order to reduce uncertainty it is important to achieve a more
efficient flow it will always be applicable to attempt to separate emergency and non-emergency
patients. The referral flow is, as mentioned above, central in the regular operations for all
ambulatory care service units. Referrals should also be possible to control and manage to greater
extent than patients since they are not subject to individual preferences. This gives the potential
for short term wins with widespread positive effects.
We also think that there is great potential in planning for events that have the potential to disrupt
the regular operations. As noted above, it is not possible to anticipate everything that happens in
advance, especially in a department that handles emergency patients, but it should be possible to
plan for certain scenarios that might occur. An ambulatory care service provider should therefore
focus on trying to understand all contingencies and events that might inhibit and disrupt the regular
operations.
70
7.1.1 MAIN RESEARCH QUESTION
The discussion on the sub-questions finally leads back to our main research question: “How can
production planning and control aid and improve the work process of an ambulatory care service
provider in a major emergency hospital in Sweden?”
We think that the application and utilisation of production planning and control methods and
principles show promise for ambulatory care services in Sweden. Managing the quality of
ambulatory care services is the key in fulfilling the purposes in for example diagnostics. Production
planning and control may also result in more predictable actions from the referring department’s
perspective. This in turn enables an organisation to optimise their operations further.
We conclude that successful implementation of production planning and control in the operations
of an ambulatory care service provider likely needs to be permeated by a few underlying principles.
First, the implementation should begin with a deep understanding of the process and the flows in
the unit. A mapping of the process will facilitate an interpretation and analysis that aim to promote
both operational and medical objectives. This will aid in the inclusion of production planning and
control in the clinical work, and not as separate activities performed by management.
Second, the production planning and control need to be based and adapted to the external
environment the unit is acting in, including factors such as referring departments, patient mix and
flows. This will enable extending partnerships and collaboration to facilitate mutual benefits and
also enable a more grounded planning. This requires a special emphasis on the analysis and
collection of quantitative data rather than cumulative personal experiences.
Third, continuous feedback on performance measurements devised on a department level is a key
to sustainable success. The continuous feedback really is vital to promoting continuous
improvements as well as in facilitating more long term planning and a more dynamic adaptation
of operations if needed.
7.2 CONCLUDING REMARKS
These concluding remarks aim to give an indication as to how the thesis can serve as a foundation
for further studies and also attempt to put the thesis in a broader perspective and reflect on the
sustainability of the study.
We also discuss what we think that the radiology department at Danderyds sjukhus should focus
on next in order to improve and to remedy the identified issues prior to the study. These
recommendations could not be part of the main conclusions to the study, as the findings are of
quite specific nature and contingent on unique aspects of the setting that could not be further
generalised to pertain to any ambulatory care service provider.
7.2.1 RECOMMENDATIONS TO THE RADIOLOGY DEPARTMENT AT DANDERYDS SJUKHUS
We have concluded that ambulatory care services in general are suitable for the implementation
of various production planning and control techniques, methods and principles. This thesis is
however based on an extensive case study in the radiology department at Danderyds sjukhus. We
have also concluded that in order for an implementation of production planning and control
initiative to be successful it is vital that they are adapted and implemented with special care taken
to the setting in which the unit is acting in. This study has therefore been complemented with more
specific practical recommendations in the radiology department at Danderyds sjukhus that are
adapted to the specific settings of the department.
71
We think that the next steps for the radiology department at Danderyds sjukhus should be:
Institute a continuous mapping and tracking of demand. The understanding of expected
demand is based on accumulated experiences in the department. We also think that a data
driven approach can incorporate changes in the environment quicker since it would shorten
the feedback loop from the department.
Formulate goals and performance measurements that are suitable for continuous
feedback. The department could use more specific goals that reconcile both operational and
medical objectives. Inspiration can be found in the radiology department at
“Universitetssjukhuset i Linköping”. These goals and performance measurements should
be subject to continuous follow-up and feedback in the running operations.
Formalise collaboration with other departments. We have seen that interdepartmental
collaboration works best when it can be systematised and is built on personal connections.
We therefore believe that the radiology department at Danderyds sjukhus should focus on
mapping and formalising interdepartmental. Not only physicians should be involved in
these collaborations; operational matter such as logistics solutions can for example be
developed by nurses since they are more involved in these matters than physicians.
Separate emergency and non-emergency flows. The separation of emergency and non-
emergency examinations shows promise. If the department were to separate the two flows
and specialise machines it would foremost enable a more efficient handling of non-
emergency patients.
7.2.2 REFLECTIONS ON SUSTAINABILITY
When speaking of sustainability in research, the term sustainable development is an
interchangeable term. For an engineer in today’s society it is vital to reflect upon the sustainable
development of the research one conducts. This has grown to be an important topic from the
growing awareness in society about environmental and socio-economic issues (Hopwood et al.,
2005). The exact definition of sustainability differ depending on source but a widely quoted
definition from WCED (1987) have sustainable development as “…development that meets the
needs of the present without compromising the ability of future generations to meet their own
needs” (Faezipour and Ferreira, 2011). This challenges the post-war idea that international
prosperity and human well-being comes from global trade and industry. This idea still heavily
influences economic policy on a global level (Hopwood et al., 2005).
WCED mentions three pillars of sustainable development: social, economic and environmental
(Faezipour and Ferreira, 2011). According to definitions from the UN, the social pillar regards
civil and political rights as well as equal access to resources, wealth and opportunities and the
well-being of people. The economic pillar meanwhile regards topics such as economic security,
access to an adequate livelihood. Also included is the fair distribution and efficient allocation of
resources. Last, the environmental pillar includes ecological concerns and processes that support
sustainability in areas such as recycling, waste management and renewable energy. These are all
topics that relate to protection of the earth and its resources (Faezipour and Ferreira, 2011).
For this thesis, environmental sustainability has not been a big concern and the outcomes lack a
discussion about the environment. This is due to it not being deemed interesting for us regarding
the case. Healthcare is vital in a society and we think that offering the best care possible should be
deemed more important than environmental impact. It is possible to discuss the environmental
impact of healthcare and issues such as release of hospital.
72
As discussed in the introduction of this thesis, the demand for healthcare is increasing with an
aging population. Since the resource are not infinite there has to be development regarding
efficiency, which further stresses the importance of writing a thesis studying the topic we have
studied. Economic and social sustainability is therefore of high importance when it comes to
healthcare, and both the needs of today and the future should be regarded when developing
healthcare sustainably (Faezipour and Ferreira, 2011). The main sustainability concerns taken in
this thesis has been to emphasise more efficient resource utilisation and to not promote acquisition
of resources that were not utilised beforehand.
We would argue that this thesis provides ideas of benefit for social development by enabling more
efficient ways to provide healthcare. This would mean that better care is offered to a lower cost,
which may not have been proven here but a foundation for future development has been laid.
7.2.3 SUGGESTIONS FOR FURTHER RESEARCH
We think that there are a number of interesting areas of further research that might shed light on
the topic of production planning and control in the setting of a radiology department.
One issue to explore could be: How to standardise measuring quality in radiology: which quality
measures are most applicable? Today, there are no standard in measuring for radiology and in
Sweden policy makers have not found a suitable way to monitor radiology on a national level.
Another topic that we found interesting after conducting this study is: What will be the impact of
the “Standardiserade vårdförlopp” in Sweden? This is a new idea first developed for certain kinds
of cancer. The care given is specified centrally and it should be the same anywhere in Sweden
(Regionala cancercentrum i samverkan – landstingens och regionernas nationella samverkan inom
cancersjukvården, 2015). The effects of this may be an increase in standardisation overall, if it
proves good and this will be an interesting topic to study. Out of our findings and the literature in
the field of production planning and control we argue that there is room for increase of
standardisation in healthcare operations.
A third topic could be to elaborate upon tasks that could be transferred from physicians to nurses.
This has started to some extent but if it is possible to increase the amount of tasks conducted by
nurses, hospitals could save money and physicians could be used only where their skills are really
needed.
73
8 BIBLIOGRAPHY Ahlstrom, P., 2004, Lean service operations: translating lean production principles to service
operations: International Journal of Services Technology and Management, v. 5, p. 545-564.
Akkermans, H. A., and L. N. Van Wassenhove, 2013, Searching for the grey swans: the next 50 years of production research: International Journal of Production Research, v. 51, p. 6746-6755.
Andersen, H., K. A. Røvik, and T. Ingebrigtsen, 2014, Lean thinking in hospitals: is there a cure for the absence of evidence? A systematic review of reviews: BMJ open, v. 4, p. e003873.
Andersson, M., 2014, Budgetpropositionen för 2015, in Finansdepartementet, ed., Stockholm, Sweden, Regeringen.
Argote, L., and P. Ingram, 2000, Knowledge transfer: A basis for competitive advantage in firms: Organizational behavior and human decision processes, v. 82, p. 150-169.
Arnaboldi, M., and G. Azzone, 2010, Constructing performance measurement in the public sector: Critical Perspectives on Accounting, v. 21, p. 266-282.
Axelsson, M., 2015, Region Östergötland - Röntgenkliniken i Linköping. Beliën, J., and E. Demeulemeester, 2008, A branch-and-price approach for integrating nurse
and surgery scheduling: European journal of operational research, v. 189, p. 652-668. Bhasin, S., 2008, Lean and performance measurement: Journal of Manufacturing Technology
Management, v. 19, p. 670-684. Binkhuysen, F. B., and E. Ranschaert, 2011, Teleradiology: evolution and concepts: European
journal of radiology, v. 78, p. 205-209. Blackmore, C. C., 2007, Defining quality in radiology: Journal of the American College of
Radiology, v. 4, p. 217-223. Blake, J. T., and M. W. Carter, 2002, A goal programming approach to strategic resource
allocation in acute care hospitals: European Journal of Operational Research, v. 140, p. 541-561.
Bowerman, J., and D. Fillingham, 2007, Can lean save lives?: Leadership in Health Services, v. 20, p. 231-241.
Brailsford, S., and J. Vissers, 2011, OR in healthcare: A European perspective: European journal of operational research, v. 212, p. 223-234.
Britten, N., 1995, Qualitative research: qualitative interviews in medical research: Bmj, v. 311, p. 251-253.
Burgess, N., and Z. Radnor, 2013, Evaluating Lean in healthcare: International journal of health care quality assurance, v. 26, p. 220-235.
Buzacott, J. A., H. Corsten, R. Gössinger, and H. M. Schneider, 2012, Production Planning and Control: Basics and Concepts, Oldenbourg Verlag.
Cameron, M., S. Cranfield, V. Iles, and J. Stone, 2001, Making Informed Decisions On Change - Key Points for Health Care Managers And Professionals, London, NHS Service Delivery and Organisation R&D Programme.
Center for Devices and Radiological Health, 2015a, Medical Imaging - MRI (Magnetic Resonance Imaging).
Center for Devices and Radiological Health, 2015b, Medical Imaging - Ultrasound Imaging. Center for Devices and Radiological Health, 2015c, Medical X-ray Imaging - Computed
Tomography (CT). Center for Devices and Radiological Health, 2015d, Medical X-ray Imaging - Fluoroscopy. Center for Devices and Radiological Health, 2015e, Medical X-ray Imaging - Radiography. Chan, S., 2002, The Importance fof Strategy for the Evolving Field of Radiology, New York,
Radiological Society of North America, Department of Radiology Columbia University.
Choy, G., and R. A. Novelline, 2013, Past, present, and future of emergency radiology: Canadian Association of Radiologists Journal, v. 64, p. 85-89.
Christopher, M., 2012, Logistics and supply chain management, Pearson UK. Ciemnoczolowski, D. D., and Y. A. Bozer, 2013, Performance evaluation of small -batch
container delivery systems used in lean manufacturing–Part 2: number of Kanban and workstation starvation: International Journal of Production Research, v. 51, p. 568-581.
74
Clegg, J. M. a. P. D., Ben, M. Elg, K. Palmberg Broryd, and B. Kollberg, 2013, Performance measurement to drive improvements in healthcare practice: International Journal of Operations & Production Management, v. 33, p. 1623-1651.
Collis, J., and R. Hussey, 2014, Business Research - a practical guide for undergraduate and postgraduate students: New York, New York, USA, PALGRAVE MACMILLAN HIGHER EDUCATION.
Conti, R., J. Angelis, C. Cooper, B. Faragher, and C. Gill, 2006, The effects of lean production on worker job stress: International journal of operations & production management, v. 26, p. 1013-1038.
Cook, T., P. Hunt, and R. Hoppman, 2007, Emergency Medicine Leads the Way for Training Medical Students in Clinician‐based Ultrasound: A Radical Paradigm Shift in Patient Imaging: Academic Emergency Medicine, v. 14, p. 558-561.
Danderyds sjukhus AB, 2015a, Bild och funktion - Danderyds sjukhus, Danderyds sjukhus AB.
Danderyds sjukhus AB, 2015b, Fakta Danderyds sjukhus ur bokslut 2013.pdf, http://www.danderydssjukhus.se/Global/Om%20sjukhuset/Organisation/Dokument/Fakta%20Danderyds%20sjukhus%20ur%20bokslut%202013.pdf, Danderyds sjukhus, p. Fakta om Danderyds sjukhus.
Danderyds sjukhus AB, 2015c, Lean - Danderyds sjukhus, http://www.danderydssjukhus.se/Om-oss/Kvalitetsarbete/Lean/, Danderyds sjukhus AB.
De Souza, L. B., 2009, Trends and approaches in lean healthcare: Leadership in Health Services, v. 22, p. 121-139.
De Vries, G., J. Bertrand, and J. Vissers, 1999, Design requirements for health care production control systems: Production Planning & Control, v. 10, p. 559-569.
Denscombe, M., 2004, Forskningens grundregler, v. 1:5: Lund, Studentlitteratur AB. Devaraj, S., T. T. Ow, and R. Kohli, 2013, Examining the impact of information technology
and patient flow on healthcare performance: A Theory of Swift and Even Flow (TSEF) perspective: Journal of Operations Management, v. 31, p. 181-192.
Dubois, A., and L.-E. Gadde, 2002, Systematic combining: an abductive approach to case research: Journal of business research, v. 55, p. 553-560.
Dunleavy, P., H. Margetts, S. Bastow, and J. Tinkler, 2006, New public management is dead—long live digital-era governance: Journal of public administration research and theory, v. 16, p. 467-494.
Ebrahimpour, M., and B. Modarress Fathi, 1985, Dynamic simulation of a Kanban production inventory system: International Journal of Operations & Production Management, v. 5, p. 5-14.
Eccles, R. G., 1990, The performance measurement manifesto: Harvard business review, v. 69, p. 131-137.
Eisenhardt, K. M., 1989, Building theories from case study research: Academy of management review, v. 14, p. 532-550.
Faezipour, M., and S. Ferreira, 2011, Applying systems thinking to assess sustainabilit y in healthcare system of systems: International Journal of System of Systems Engineering, v. 2, p. 290-308.
Feilzer, M. Y., 2010, Doing mixed methods research pragmatically: Implications for the rediscovery of pragmatism as a research paradigm: Journal of mixed methods research, v. 4, p. 6-16.
Francis, J. J., M. Johnston, C. Robertson, L. Glidewell, V. Entwistle, M. P. Eccles, and J. M. Grimshaw, 2010, What is an adequate sample size? Operationalising data saturation for theory-based interview studies: Psychology and Health, v. 25, p. 1229-1245.
Gelders, L. F., and L. N. Van Wassenhove, 1981, Production planning: a review: European Journal of Operational Research, v. 7, p. 101-110.
Georgiadis, P., and C. Michaloudis, 2012, Real-time production planning and control system for job-shop manufacturing: A system dynamics analysis: European Journal of Operational Research, v. 216, p. 94-104.
Gibbert, M., W. Ruigrok, and B. Wicki, 2008, Reserach notes and commentaries what passes as a rigorous case study?: Strategic Management Journal, v. 29, p. 1465-1474.
Gill, I. E., S. Ondategui-Parra, E. Nathanson, J. Seiferth, and P. R. Ros, 2005, Strategic planning in radiology: Journal of the American College of Radiology, v. 2, p. 348-357.
75
Goddard, M., R. Mannion, and P. Smith, 2004, Enhancing performance in health care: a theoretical perspective on agency and the role of information: Studying the organisation and delivery of health services: a reader, v. 9, p. 199.
Goldratt, E. M., 1990, Theory of constraints, North River Croton-on-Hudson, NY. Grufman, C., and D. Sundblad, 2005, Kartläggning och analys av patientflöde vid
Södersjukhuset. Gupta, M. C., and L. H. Boyd, 2008, Theory of constraints: a theory for operations
management: International Journal of Operations & Production Management, v. 28, p. 991-1012.
Hall, R., D. Belson, P. Murali, and M. Dessouky, 2006, Modeling patient flows through the healthcare system, Patient flow: Reducing delay in healthcare delivery, Springer, p. 1-44.
Hans, E. W., M. Van Houdenhoven, and P. J. Hulshof, 2012, A framework for healthcare planning and control, Handbook of healthcare system scheduling, Springer, p. 303-320.
Harper, P. R., 2002, A framework for operational modelling of hospital resources: Health care management science, v. 5, p. 165-173.
Holweg, M., 2007, The genealogy of lean production: Journal of operations management, v. 25, p. 420-437.
Hopwood, B., M. Mellor, and G. O'Brien, 2005, Sustainable development: mapping different approaches: Sustainable development, v. 13, p. 38-52.
Hsieh, H.-F., and S. E. Shannon, 2005, Three approaches to qualitative content analysis: Qualitative health research, v. 15, p. 1277-1288.
Hulshof, P. J., N. Kortbeek, R. J. Boucherie, E. W. Hans, and P. J. Bakker, 2012, Taxonomic classification of planning decisions in health care: a structured review of the state of the art in OR/MS: Health systems, v. 1, p. 129-175.
Jacobson, S., L. Martin, and Å. Lindström Hammar, 2015, Kvalitets- och patientsäkerhetsberättelse 2013 för webben.pdf, www.danderydssjukhus.se, Enheten för kommunikation Danderyds Sjukhus.
Jimenez, C. H. O., J. A. Machuca, P. Garrido-Vega, and R. Filippini, 2014, The pursuit of responsiveness in production environments: From flexibility to reconfigurability: International Journal of Production Economics.
Jonsson, P., and S.-A. Mattsson, 2003, The implications of fit between planning environments and manufacturing planning and control methods: International Journal of Operations & Production Management, v. 23, p. 872-900.
Joosten, T., I. Bongers, and R. Janssen, 2009, Application of lean thinking to health care: issues and observations: International Journal for Quality in Health Care, v. 21, p. 341-347.
Khan, S., 2013, Measuring radiologist workload: past, present and future: Diagnostic Imaging Europe, v. 30, p. 27-29.
Kilgore, R. V., and R. W. Langford, 2009, Reducing the failure risk of interdisciplinary healthcare teams: Critical care nursing quarterly, v. 32, p. 81-88.
Kopach-Konrad, R., M. Lawley, M. Criswell, I. Hasan, S. Chakraborty, J. Pekny, and B. N. Doebbeling, 2007, Applying systems engineering principles in improving health care delivery: Journal of general internal medicine, v. 22, p. 431-437.
Kruskal, J. B., A. Reedy, L. Pascal, M. P. Rosen, and P. M. Boiselle, 2012, Quality Initiatives Lean Approach to Improving Performance and Efficiency in a Radiology Department: Radiographics, v. 32, p. 573-588.
Köchel, P., and U. Nieländer, 2002, Kanban optimization by simulation and evolution: Production Planning & Control, v. 13, p. 725-734.
Martinez‐Moyano, I. J., and G. P. Richardson, 2013, Best practices in system dynamics modeling: System Dynamics Review, v. 29, p. 102-123.
Mazzocato, P., C. Savage, M. Brommels, H. Aronsson, and J. Thor, 2010, Lean thinking in healthcare: a realist review of the literature: Quality and Safety in Health Care, v. 19, p. 376-382.
McKinsey & Co, 2013, Genomlysning av Stockholms akutmottagningar, SLL. Melnyk, S. A., U. Bititci, K. Platts, J. Tobias, and B. Andersen, 2014, Is performance
measurement and management fit for the future?: Management Accounting Research, v. 25, p. 173-186.
Micheli, P., and L. Mari, 2014, The theory and practice of performance measurement: Management Accounting Research, v. 25, p. 147-156.
76
Middleton, P., and D. Joyce, 2012, Lean software management: BBC Worldwide case study: Engineering Management, IEEE Transactions on, v. 59, p. 20-32.
Molin, R., and L. Eckerström, 2008, The Swedish Healthcare System: How Does It Compare to Other EU countries, the United States and Norway?
Mourtzis, D., and M. Doukas, 2014, Design and planning of manufacturing networks for mass customisation and personalisation: Challenges and Outlook: Procedia CIRP, v. 19, p. 1-13.
Müller, E., 2006, Production planning and operation in competence-cell-based networks: Production Planning & Control, v. 17, p. 99-112.
Nguyen, V. T., A. Friis Sommer, K. Steger-Jensen, and H.-H. Hvolby, 2014, The Misalignment between Hospital Planning Frameworks and their Planning Environment - A conceptual Matchin Approach: Advances in Production Management Systems, p. 675-682.
NHS Choices, 2015, Angiography - Complications - NHS Choices. Olaitan, O., A. Rotondo, P. Young, and J. Geraghty, 2014, On Designing Robust Kanban
Production Control Strategies in Multiproduct Manufacturing Environments: Handbook of Research on Design and Management of Lean Production Systems, p. 68.
Olhager, J., 2013, Evolution of operations planning and control: from production to supply chains: International Journal of Production Research, v. 51, p. 6836-6843.
Ozcan, Y. A., and J. S. Legg, 2014, Performance measurement for radiology providers: a national study: International Journal of Healthcare Technology and Management, v. 14, p. 209-221.
Persona, A., D. Battini, and C. Rafele, 2008, Hospital efficiency management: the just-in-time and Kanban technique: International Journal of Healthcare Technology and Management, v. 9, p. 373-391.
Pollitt, C., S. Van Thiel, and V. Homburg, 2007, New public management in Europe: Management online review, p. 1-6.
Prencipe, A., and F. Tell, 2001, Inter-project learning: processes and outcomes of knowledge codification in project-based firms: Research policy, v. 30, p. 1373-1394.
Radnor, Z. J., M. Holweg, and J. Waring, 2012, Lean in healthcare: the unfilled promise?: Social Science & Medicine, v. 74, p. 364-371.
Rao, V. M., and D. C. Levin, 2011, The value-added services of hospital-based radiology groups: Journal of the American College of Radiology, v. 8, p. 626-630.
Regionala cancercentrum i samverkan – landstingens och regionernas nationella samverkan inom cancersjukvården, 2015, Standardiserade vårdförlopp i cancervården.
Roth, A. V., and R. Dierdonck, 1995, Hospital resource planning: concepts, feasibility, and framework: Production and operations management, v. 4, p. 2-29.
Simpson, J., and F. Falter, 2014, The Ideal Radiology Suite, Imaging the ICU Patient, Springer, p. 79-83.
Slack, N., S. Chambers, and R. Johnston, 2010, Operations management, Pearson Education. Sugimori, Y., K. Kusunoki, F. Cho, and S. Uchikawa, 1977, Toyota production system and
kanban system materialization of just-in-time and respect-for-human system: The International Journal of Production Research, v. 15, p. 553-564.
Sunshine, J. H., D. R. Hughes, C. Meghea, and M. Bhargavan, 2010, What factors affect the productivity and efficiency of physician practices?: Medical care, v. 48, p. 110-117.
Tako, A. A., and S. Robinson, 2012, The application of discrete event simulation and system dynamics in the logistics and supply chain context: Decision support systems, v. 52, p. 802-815.
Taylor Frederick, W., 1911, The principles of scientific management: New York & London: Harper Brothers.
Testi, A., E. Tanfani, and G. Torre, 2007, A three-phase approach for operating theatre schedules: Health Care Management Science, v. 10, p. 163-172.
Tonelli, F., M. Paolucci, D. Anghinolfi, and P. Taticchi, 2013, Production planning of mixed-model assembly lines: a heuristic mixed integer programming based approach: Production Planning & Control, v. 24, p. 110-127.
Turner, R., D. Ingold, J. A. Lane, R. Madachy, and D. Anderson, 2012, Effectiveness of kanban approaches in systems engineering within rapid response environments: Procedia Computer Science, v. 8, p. 309-314.
77
Umble, E. J., R. R. Haft, and M. M. Umble, 2003, Enterprise resource planning: Implementation procedures and critical success factors: European journal of operational research, v. 146, p. 241-257.
Van Den Berg, J. P., 1999, A literature survey on planning and control of warehousing systems: IIE transactions, v. 31, p. 751-762.
Van der Meer-Kooistra, J., and E. G. Vosselman, 2000, Management control of interfirm transactional relationships: the case of industrial renovation and maintenance: Accounting, organizations and society, v. 25, p. 51-77.
van Merode, G. G., S. Groothuis, and A. Hasman, 2004, Enterprise resource planning for hospitals: International Journal of Medical Informatics, v. 73, p. 493-501.
Verčič, A. T., D. Verčič, and K. Sriramesh, 2012, Internal communica tion: Definition, parameters, and the future: Public relations review, v. 38, p. 223-230.
Vissers, J. M., J. Bertrand, and G. De Vries, 2001, A framework for production control in health care organizations: Production Planning & Control, v. 12, p. 591-604.
Vlachos, D., P. Georgiadis, and E. Iakovou, 2007, A system dynamics model for dynamic capacity planning of remanufacturing in closed-loop supply chains: Computers & Operations Research, v. 34, p. 367-394.
Vos, L., S. Groothuis, and G. G. van Merode, 2007, Evaluating hospital design from an operations management perspective: Health care management science, v. 10, p. 357-364.
Wang, J., S. Quan, J. Li, and A. M. Hollis, 2012, Modeling and analysis of work flow and staffing level in a computed tomography division of University of Wisconsin Medical Foundation: Health care management science, v. 15, p. 108-120.
Waring, J. J., and S. Bishop, 2010, Lean healthcare: Rhetoric, ritual and resistance: Social Science & Medicine, v. 71, p. 1332-1340.
Womack, J. P., and D. T. Jones, 2010, Lean thinking: banish waste and create wealth in your corporation, Simon and Schuster.
World Health Organization, 1998, Training in diagnostic ultrasound: essentials, principles and standards: report of a WHO study group.
Yin, R. K., 1999, Enhancing the quality of case studies in health services research: Health Services Research, v. 34, p. 1209.
Yin, R. K., 2009, Case Study Research. Design and Methods: Thousand Oaks, Sage Publications.
Zijm, W., 2000, Towards intelligent manufacturing planning and control systems: OR-Spektrum, v. 22, p. 313-345.